WEBVTT

00:00:00.000 --> 00:00:03.477
Good morning. >> [applause]

00:00:03.560 --> 00:00:05.637
>> Morning.

00:00:05.720 --> 00:00:08.237
It's uh fantastic

00:00:08.320 --> 00:00:11.917
to be back uh in San Francisco at Build.

00:00:12.000 --> 00:00:16.917
It's always fun to be at developer
conferences at times of great change.

00:00:17.000 --> 00:00:20.797
You know, developer conferences are
always about understanding tech shifts,

00:00:20.880 --> 00:00:23.477
understanding the new stack, but it's

00:00:23.560 --> 00:00:26.677
also about really coming to grips with

00:00:26.760 --> 00:00:29.277
the new opportunity, right? For us as

00:00:29.360 --> 00:00:31.557
developers, for the companies that we

00:00:31.640 --> 00:00:34.717
work at, as well as the broader world.

00:00:34.800 --> 00:00:37.157
Uh so, today we're going to unpack this.

00:00:37.240 --> 00:00:38.957
This conference is all about that. And

00:00:39.040 --> 00:00:41.917
if there's one key takeaway,

00:00:42.000 --> 00:00:44.957
it would be this. How do you all

00:00:45.040 --> 00:00:48.637
participate fully in this frontier

00:00:48.720 --> 00:00:53.237
intelligence ecosystem? Right? It's not
about any one piece of technology that

00:00:53.320 --> 00:00:55.597
you'll hear about or even the platform

00:00:55.680 --> 00:00:59.317
itself. It's about the value that you

00:00:59.400 --> 00:01:04.237
can build, you can compound, you can
create on top of the platform. That's

00:01:04.320 --> 00:01:06.197
what a developer conference needs to be

00:01:06.280 --> 00:01:08.637
about, and that's what we will focus on.

00:01:08.720 --> 00:01:11.677
So, let's kind of get into it. Um let's

00:01:11.760 --> 00:01:14.677
take a look at this AI stack that this

00:01:14.760 --> 00:01:19.837
conference is going to unpack in a great
amount of detail. It starts, of course,

00:01:19.920 --> 00:01:21.477
with the compute fabric, right? That

00:01:21.560 --> 00:01:25.117
ubiquitous compute fabric that is spans

00:01:25.200 --> 00:01:29.957
the edge and the cloud.
Uh you then have this layer, which is

00:01:30.040 --> 00:01:31.677
the emerging layer where you have the

00:01:31.760 --> 00:01:34.637
models, you have context, you have the

00:01:34.720 --> 00:01:38.077
tools the models can access. Um and then

00:01:38.160 --> 00:01:40.237
on top of that, you have the runtime

00:01:40.320 --> 00:01:45.757
where you deploy your agents and
applications that you built uh on top of

00:01:45.840 --> 00:01:49.437
uh the context layer, on top of the
model layer. And then, of course, you

00:01:49.520 --> 00:01:53.837
have the best tooling uh to do all
of this. And then, you have the security,

00:01:53.920 --> 00:01:58.597
the compliance, uh and the governance,
right? That's the simplest form of the

00:01:58.680 --> 00:02:02.597
tech stack. But let's kind
of start where it always

00:02:02.680 --> 00:02:04.917
starts, which is infrastructure.

00:02:05.000 --> 00:02:07.357
And in fact, let's start at the edge

00:02:07.440 --> 00:02:09.157
with Windows.

00:02:09.240 --> 00:02:12.117
Because when you step back,

00:02:12.200 --> 00:02:15.437
the amount of compute there is at the

00:02:15.520 --> 00:02:17.477
edge is actually astounding, right? I

00:02:17.560 --> 00:02:21.677
mean, think about every NPU, GPU, CPU

00:02:21.760 --> 00:02:24.957
even. Um, you every PC, if you sort of

00:02:25.040 --> 00:02:27.237
aggregate that, that's a lot of compute

00:02:27.320 --> 00:02:30.677
power. So, we asked ourselves one simple

00:02:30.760 --> 00:02:32.117
question.

00:02:32.200 --> 00:02:34.597
Right? If we can deliver

00:02:34.680 --> 00:02:38.197
unmetered intelligence to every desk and

00:02:38.280 --> 00:02:42.077
every home, right? It takes us all
the way back to the very beginning, but

00:02:42.160 --> 00:02:46.637
that's what we said.
Can we do that in this era of AI?

00:02:46.720 --> 00:02:50.677
And in fact, we're in some sense
delivering that already today, right?

00:02:50.760 --> 00:02:52.317
When you look at something like, say,

00:02:52.400 --> 00:02:54.597
Outlook summarize, it's actually using

00:02:54.680 --> 00:02:57.957
onboard AI locally. Uh, same thing with

00:02:58.040 --> 00:03:01.077
PowerPoint all text, uh, or Teams super

00:03:01.160 --> 00:03:03.477
resolution. Uh, it's just not Microsoft

00:03:03.560 --> 00:03:05.237
stuff. It's a bit For example, Adobe

00:03:05.320 --> 00:03:08.357
After Effects or Premiere are both using

00:03:08.440 --> 00:03:11.717
Windows ML, um, across, uh, NPUs and

00:03:11.800 --> 00:03:15.917
GPUs for local processing. And so,
today one of the things that I'm

00:03:16.000 --> 00:03:17.997
really excited about in order to tap

00:03:18.080 --> 00:03:21.437
into all this compute power is to expand

00:03:21.520 --> 00:03:25.077
the scope of Windows ML and Windows AI,

00:03:25.160 --> 00:03:30.437
right? So, you now have the full install
base of GPUs, uh, that you can get to.

00:03:30.520 --> 00:03:34.157
So, I'm really thrilled that every
developer out there can count on

00:03:34.240 --> 00:03:39.437
building for local onboard AI, uh, and
then have it run across all of, uh, the

00:03:39.520 --> 00:03:43.197
install base. Now, if you

00:03:43.280 --> 00:03:45.757
We are also announcing two very cool new

00:03:45.840 --> 00:03:48.877
models, uh, that are all going to run on

00:03:48.960 --> 00:03:52.117
Windows, uh, inbox. Uh, the first is a

00:03:52.200 --> 00:03:55.957
new SLM. It's sort of a more efficient um

00:03:56.040 --> 00:04:00.997
model ion instruct um and it's
a it's a great uh reasoning

00:04:01.080 --> 00:04:05.757
model. And then we have the planning
model um ion plan, which is a local

00:04:05.840 --> 00:04:10.197
agentic loop. I mean, think about it,
right? You now have a full local agentic

00:04:10.280 --> 00:04:12.637
loop. You can give it tools access um

00:04:12.720 --> 00:04:15.837
and build a fully onboard agent agentic

00:04:15.920 --> 00:04:18.277
applications uh without having having to

00:04:18.360 --> 00:04:20.557
round trip to the cloud.

00:04:20.640 --> 00:04:23.157
Uh and of course, we said in order to

00:04:23.240 --> 00:04:26.677
really push the limits, realize this

00:04:26.760 --> 00:04:31.157
unmetered intelligence, there's also
a lot of hardware that's evolving, too.

00:04:31.240 --> 00:04:34.957
And so we're thrilled to see the
innovation across the Windows ecosystem,

00:04:35.040 --> 00:04:37.677
right? It's AMD with their Ryzen

00:04:37.760 --> 00:04:42.757
processors, the Panther Lake stuff from
Intel is just exciting to see. Qualcomm

00:04:42.840 --> 00:04:46.717
uh announced two sets of things,
which are great. One on the high end with

00:04:46.800 --> 00:04:49.077
Snapdragon X2 Elite.

00:04:49.160 --> 00:04:50.797
Uh and then on the lower end, uh even

00:04:50.880 --> 00:04:54.517
for sub 500 uh PCs with Snapdragon C.

00:04:54.600 --> 00:04:56.717
So, it's great to see all of this.

00:04:56.800 --> 00:04:58.917
And of course, that brings us

00:04:59.000 --> 00:05:02.517
uh to Nvidia uh and RTX Spark. Uh this

00:05:02.600 --> 00:05:05.197
is a next-generation SOC for PCs. It

00:05:05.280 --> 00:05:08.917
brings together the CPU, the GPU, as

00:05:09.000 --> 00:05:12.917
well as the AI capabilities into
a single SOC. And that's the exciting

00:05:13.000 --> 00:05:15.757
part. And it includes unified memory uh

00:05:15.840 --> 00:05:19.477
architecture and also the integrated uh

00:05:19.560 --> 00:05:24.597
DRTM, right? So, you now have all of
the systems innovation and the SOC coming

00:05:24.680 --> 00:05:27.997
together. Um and we're
really thrilled that one of

00:05:28.080 --> 00:05:31.597
the first devices that we built uh was

00:05:31.680 --> 00:05:33.957
the Surface Ultra, right? It's uh it's a

00:05:34.040 --> 00:05:37.037
beautiful uh device. And uh it really

00:05:37.120 --> 00:05:40.397
brings the power of Nvidia together with

00:05:40.480 --> 00:05:42.757
the design and the craftsmanship of

00:05:42.840 --> 00:05:46.557
Surface. It's 128 GB of uh unified

00:05:46.640 --> 00:05:50.837
memory. It's beautiful, too. Uh 2,000 and

00:05:50.920 --> 00:05:55.917
and and it has an all day
battery life. And so we're

00:05:56.000 --> 00:05:58.997
really excited to see
this later this fall.

00:05:59.080 --> 00:06:01.677
And it's also wonderful to see all the

00:06:01.760 --> 00:06:04.997
designs from all the OEM partners who

00:06:05.080 --> 00:06:09.677
are building taking advantage
of this new SOC and the new platform

00:06:09.760 --> 00:06:12.877
and bringing out some
pretty exciting machines

00:06:12.960 --> 00:06:16.797
that's going to be all
available come this fall.

00:06:16.880 --> 00:06:19.717
And so of course we said, "Okay, this is

00:06:19.800 --> 00:06:22.797
fantastic. What can we do next?" So we

00:06:22.880 --> 00:06:26.477
said, "Let's try and push this and push

00:06:26.560 --> 00:06:29.877
the architecture to its limit for

00:06:29.960 --> 00:06:34.197
developers. Right?
What if we could just max the compute,

00:06:34.280 --> 00:06:38.917
max the memory, build out
that developer machine that's

00:06:39.000 --> 00:06:40.477
the dream machine.

00:06:40.560 --> 00:06:42.997
And that's what we're announcing today.

00:06:43.080 --> 00:06:48.920
Surface RTX Spark Dev Box.
Let's roll the video.

00:06:50.690 --> 00:06:52.710
>> [music]

00:06:57.890 --> 00:06:59.910
[music]

00:07:11.650 --> 00:07:13.670
[music]

00:07:16.230 --> 00:07:18.250
[music]

00:07:21.480 --> 00:07:23.500
[music]

00:07:24.100 --> 00:07:26.120
[singing]

00:07:29.330 --> 00:07:31.350
[music]

00:07:38.500 --> 00:07:40.520
[music]

00:07:43.090 --> 00:07:48.380
[music] >> All right. >> [applause]

00:07:50.960 --> 00:07:53.117
>> Yeah, this is it's it's truly a dream

00:07:53.200 --> 00:07:56.437
machine. It's got a one petaflop of AI

00:07:56.520 --> 00:08:01.077
compute, 20 CPU cores.

00:08:01.160 --> 00:08:04.077
All of those things have 128

00:08:04.160 --> 00:08:07.717
GB of unified memory access. So super

00:08:07.800 --> 00:08:11.917
excited about this coming
in the fall and you can join the

00:08:12.000 --> 00:08:13.837
wait list. I'm on the wait list as well.

00:08:13.920 --> 00:08:15.197
So

00:08:15.280 --> 00:08:17.517
we'll get there.

00:08:17.600 --> 00:08:21.317
You know, so then we said why stop

00:08:21.400 --> 00:08:26.077
there? We said what if
we did one more thing?

00:08:26.160 --> 00:08:29.117
And in fact, I think
Jensen did that already

00:08:29.200 --> 00:08:31.437
when he said Windows is coming

00:08:31.520 --> 00:08:34.877
to the DGX station. So a new I describe

00:08:34.960 --> 00:08:37.717
it as the desktop data center, which you

00:08:37.800 --> 00:08:41.117
can have right there and running

00:08:41.200 --> 00:08:44.237
a one trillion parameter model locally.

00:08:44.320 --> 00:08:47.077
I mean, just to sort of put it
in perspective, it's pretty close to what

00:08:47.160 --> 00:08:51.797
perhaps we had when we built GPT-2 5 or 3,
right? One of the first one of the

00:08:51.880 --> 00:08:55.877
first computer supercomputer. So it's
pretty crazy to think that we've come

00:08:55.960 --> 00:09:00.837
this far where you can now have
a data center on your desktop.

00:09:00.920 --> 00:09:02.877
And of course, we're extending the

00:09:02.960 --> 00:09:04.837
developer endpoints even to the cloud.

00:09:04.920 --> 00:09:08.157
Windows 365 now has the developer

00:09:08.240 --> 00:09:11.077
distribution that's optimized for

00:09:11.160 --> 00:09:15.397
developer productivity in the cloud.
So that's the Windows 365 that I use every

00:09:15.480 --> 00:09:19.317
day. And it's simply about making Windows,

00:09:19.400 --> 00:09:24.677
whether it's on the laptop, the desktop,
the cloud, whatever, the best place to

00:09:24.760 --> 00:09:28.997
build. And so to that end,
we have tons and

00:09:29.080 --> 00:09:33.357
tons of updates, right? We're starting
by the way with one of the favorite

00:09:33.440 --> 00:09:37.037
things for all of us,
which is distraction-free

00:09:37.120 --> 00:09:41.677
dev environment. Uh we're also
introducing an intelligent terminal uh

00:09:41.760 --> 00:09:45.677
which has got built-in GitHub Copilot,
right? Terminal with sort of the Copilot

00:09:45.760 --> 00:09:48.597
intelligence. Um and of course, there's

00:09:48.680 --> 00:09:51.677
lots and lots of Linux love uh across uh

00:09:51.760 --> 00:09:53.837
Windows now. So, in

00:09:53.920 --> 00:09:57.237
you have 70 plus utilities. So, grep uh

00:09:57.320 --> 00:10:00.117
in full glory is now available for

00:10:00.200 --> 00:10:02.957
regular Windows access. Uh 70 plus

00:10:03.040 --> 00:10:05.437
utilities, as I said, uh all of that is

00:10:05.520 --> 00:10:07.797
coming to Windows. Um we're also

00:10:07.880 --> 00:10:10.877
bringing uh things from the Mac that you

00:10:10.960 --> 00:10:13.797
love. Uh things like uh Starship, Z

00:10:13.880 --> 00:10:16.597
shell, uh Homebrew is going to be native

00:10:16.680 --> 00:10:19.157
on Windows as well. So, you'll be able

00:10:19.240 --> 00:10:22.957
to uh switch to Windows. Um and today uh

00:10:23.040 --> 00:10:28.797
we're announcing uh WSL containers. Uh
this is, I know, a point of pain uh when

00:10:28.880 --> 00:10:32.077
you're managing all these uh
environments, uh switching, all the

00:10:32.160 --> 00:10:34.117
mental gymnastics of trying to keep this

00:10:34.200 --> 00:10:36.597
straight. Uh so, having first-class

00:10:36.680 --> 00:10:40.277
support uh for containers uh will really

00:10:40.360 --> 00:10:42.477
uh help us uh be in the flow when you

00:10:42.560 --> 00:10:45.317
are building and deploying locally. Um

00:10:45.400 --> 00:10:49.317
and to show you all of this on our new

00:10:49.400 --> 00:10:52.477
uh Surface RTX Spark, I wanted to uh

00:10:52.560 --> 00:10:55.197
invite up on stage um

00:10:55.280 --> 00:11:00.167
Kayla to just uh take a spin through
all the dev tools. Kayla, go ahead.

00:11:00.250 --> 00:11:02.270
>> [applause]

00:11:04.830 --> 00:11:08.077
[applause] >> Our team has
been working hard to make

00:11:08.160 --> 00:11:10.317
Windows a great place for development.

00:11:10.400 --> 00:11:13.437
And today, I'm going to show you some
major improvements and the lighters you

00:11:13.520 --> 00:11:16.997
can try out today. So,
what we have here is the default

00:11:17.080 --> 00:11:19.797
experience on the Surface RTX Spark dev

00:11:19.880 --> 00:11:22.717
box. Right away, it feels calm. There's

00:11:22.800 --> 00:11:27.277
no news feed, no widgets popping up,
no notifications, and of course, we're in

00:11:27.360 --> 00:11:31.077
dark mode. I'm immediately
ready to start development.

00:11:31.160 --> 00:11:34.797
Well, there's one thing I'd like
to change. I'd like to put my taskbar on

00:11:34.880 --> 00:11:38.477
the left. So, let me jump into
my taskbar settings using the new run,

00:11:38.560 --> 00:11:42.077
which is built leveraging
the architecture of PowerToys command

00:11:42.160 --> 00:11:46.117
palette. No, no, no.
I want it all the way on the

00:11:46.200 --> 00:11:50.757
left. There [snorts] we go.
After popular demand, we're excited to

00:11:50.840 --> 00:11:57.120
announce that vertical taskbar is now
available in Windows Insider builds.

00:11:57.280 --> 00:11:59.957
The new Surface RT Spark also has a

00:12:00.040 --> 00:12:04.437
bunch of key dev tools already installed,
like Python, Node, and many

00:12:04.520 --> 00:12:09.037
more of your favorites. All of the
developer goodness, all in one file.

00:12:09.120 --> 00:12:11.837
And if you want to get the same
experience today on your device, we're

00:12:11.920 --> 00:12:14.877
making this file available
to everyone right now.

00:12:14.960 --> 00:12:18.717
We have a public repo set up with the
configuration file and instructions for

00:12:18.800 --> 00:12:22.317
how to apply it using winget configure,
which will make the adjustments to

00:12:22.400 --> 00:12:25.677
Windows and install all the tools.

00:12:25.760 --> 00:12:29.277
Now, one cool thing that I have running
here is PowerToys' new utility called

00:12:29.360 --> 00:12:34.077
Grab and Move, which lets you hold Alt
and move the window around from anywhere.

00:12:34.160 --> 00:12:38.037
Another tip is that you can enable end
task, which lets you end the process

00:12:38.120 --> 00:12:41.077
without having to open task manager.

00:12:41.160 --> 00:12:46.277
So, let me jump into my dev drive. Dev
drives run on ReFS with Defender running

00:12:46.360 --> 00:12:49.597
in async and are optimized for performance
when it comes to development

00:12:49.680 --> 00:12:51.077
scenarios.

00:12:51.160 --> 00:12:54.477
Also, File Explorer is Git aware. We got

00:12:54.560 --> 00:12:59.637
stuff like last change author name, last
change message, the status of each file,

00:12:59.720 --> 00:13:03.517
plus my favorite is that the branch
name is on the bottom left.

00:13:03.600 --> 00:13:07.957
So, now let's get started
building and open our terminal.

00:13:08.040 --> 00:13:11.757
This is an experimental experience
called intelligent terminal that makes

00:13:11.840 --> 00:13:14.597
working with agents even more seamless.

00:13:14.680 --> 00:13:17.557
When you first install intelligent
terminal, you're greeted with the option

00:13:17.640 --> 00:13:21.637
to pick your favorite agent. I'm going
to use GitHub Copilot for today, but you

00:13:21.720 --> 00:13:24.517
can use whichever agent speaks to you.

00:13:24.600 --> 00:13:28.477
Now, here in an terminal, I have
a regular terminal pane at the top and an

00:13:28.560 --> 00:13:31.797
agent that's listening on the bottom and I
can work between them while the agent

00:13:31.880 --> 00:13:35.117
helps along the way. So,
for example, here's an error being

00:13:35.200 --> 00:13:39.437
generated. My agent pane is able
to detect it and provide a fix, which is

00:13:39.520 --> 00:13:41.357
great when I don't remember the syntax,

00:13:41.440 --> 00:13:44.437
especially for something like regex.

00:13:44.520 --> 00:13:49.677
So, I'm going to work on open claw and
I've already built it using WSL container.

00:13:49.760 --> 00:13:53.477
WSL container is a native container
experience on Windows, plus it can

00:13:53.560 --> 00:13:57.757
leverage the GPU, which is perfect
for the Surface RTX Spark.

00:13:57.840 --> 00:14:00.997
It can also reference your existing
container files, just like the one in

00:14:01.080 --> 00:14:04.317
the open claw project. So,
here's one of the files open in

00:14:04.400 --> 00:14:08.117
Microsoft edit, which ships in Windows
by default and just got syntax

00:14:08.200 --> 00:14:10.757
highlighting in its latest version.

00:14:10.840 --> 00:14:13.397
And then you can also just see
your containers running with a simple

00:14:13.480 --> 00:14:18.157
container images command.
And since we're on the topic of WSL,

00:14:18.240 --> 00:14:20.077
we're providing a WSL profile that's

00:14:20.160 --> 00:14:25.197
designed to feel comfortable for those
of you who use tools like Starship, ZSH,

00:14:25.280 --> 00:14:28.877
and Homebrew. So, it comes
pre-configured with all of

00:14:28.960 --> 00:14:32.317
your favorite utilities and it's
available in the repo that I showed

00:14:32.400 --> 00:14:34.437
earlier. It also includes one of my

00:14:34.520 --> 00:14:37.437
favorites, Btop.

00:14:37.520 --> 00:14:39.757
So, the Surface RTX Spark is designed

00:14:39.840 --> 00:14:44.957
for developer heavy workloads, including
serving large local models for coding.

00:14:45.040 --> 00:14:48.997
I've already done some development with
a 120 billion parameter model that most

00:14:49.080 --> 00:14:53.117
machines can't even load. So,
here's a quick view of my usage and

00:14:53.200 --> 00:14:57.037
we can see how many tokens I've used
locally. So, we're looking at about 3.4

00:14:57.120 --> 00:15:00.557
million tokens leveraged
on the device itself.

00:15:00.640 --> 00:15:04.837
Now, we can kick off multiple sub-agents
using fleet and just so we don't have to

00:15:04.920 --> 00:15:08.557
watch me type, I'm going to use
Copilot's voice feature, which is also

00:15:08.640 --> 00:15:12.117
leveraging its own local model. So,
I'll just hold spacebar and tell it what I

00:15:12.200 --> 00:15:14.920
want it to do.

00:15:15.040 --> 00:15:19.277
Find any console.writeline
or debug.writeline calls in the tray and

00:15:19.360 --> 00:15:21.597
node projects and convert them

00:15:21.680 --> 00:15:26.717
to the standard logger used
elsewhere in the code base.

00:15:26.800 --> 00:15:28.637
Here we go. Now the main agent will

00:15:28.720 --> 00:15:33.797
delegate sub agent tasks of appropriate
complexity to the local model utilizing

00:15:33.880 --> 00:15:36.837
my GPU and making it more cost-efficient.

00:15:36.920 --> 00:15:40.717
Now as developers while we're debugging
we're often looking through log files to

00:15:40.800 --> 00:15:46.397
diagnose any issues. Sometimes finding the
location of the log files is a challenge.

00:15:46.480 --> 00:15:51.317
I'd love to be able to just type
something like grep log and find them all.

00:15:51.400 --> 00:15:55.437
Oh how sweet. So on top
of already adding curl, tar,

00:15:55.520 --> 00:15:57.637
and pseudo to Windows now we're adding

00:15:57.720 --> 00:16:00.997
over 75 command line utilities like env,

00:16:01.080 --> 00:16:07.600
head, tail, and touch for those of us
who love to live in the terminal.

00:16:08.040 --> 00:16:11.837
So I found all my log files but now
having to parse them is the second

00:16:11.920 --> 00:16:14.117
challenge. Well I've actually had Ion

00:16:14.200 --> 00:16:17.157
instruct practically performing analyses

00:16:17.240 --> 00:16:20.157
on my log files this whole time.

00:16:20.240 --> 00:16:23.557
Now I can quickly diagnose anything
that's gone wrong in my development.

00:16:23.640 --> 00:16:27.597
Plus I don't have to worry about
token usage because it's all local.

00:16:27.680 --> 00:16:31.837
We can even take a look at our machine's
resources. The models are loaded and you

00:16:31.920 --> 00:16:34.277
can see 90 gigs of RAM being utilized by

00:16:34.360 --> 00:16:37.117
the GPU truly showcasing the full power

00:16:37.200 --> 00:16:39.557
of the Surface RTX Spark.

00:16:39.640 --> 00:16:44.437
We were able to use three local models
simultaneously unmetered while going

00:16:44.520 --> 00:16:50.440
about our our regular dev flow
without a hitch. That's huge.

00:16:51.990 --> 00:16:54.010
>> [applause]

00:16:56.720 --> 00:17:00.317
>> So I know you're going to love what the
team's been working on. We hope it gives

00:17:00.400 --> 00:17:01.877
you a glimpse of what's possible on

00:17:01.960 --> 00:17:08.280
Windows today and where we're headed next.
Thank you. Back to you Satya.

00:17:09.020 --> 00:17:11.040
>> [applause]

00:17:11.400 --> 00:17:13.717
>> Yeah that's the really the beginning of

00:17:13.800 --> 00:17:15.997
this idea of unmetered intelligence

00:17:16.080 --> 00:17:18.477
having those models and having the

00:17:18.560 --> 00:17:21.717
agents using the models work in parallel

00:17:21.800 --> 00:17:25.957
to what you may be doing along with the
cloud as well. And that I think is what

00:17:26.040 --> 00:17:28.517
the platform enables as first class.

00:17:28.600 --> 00:17:33.077
Now, let's move to the cloud.
You know, the driving equation for us

00:17:33.160 --> 00:17:38.117
remains the same, which is tokens per
dollar per watt. How do we optimize

00:17:38.200 --> 00:17:41.677
around this, right? So, when we think
about the systems problem, we think

00:17:41.760 --> 00:17:44.237
about electrons coming on one end and

00:17:44.320 --> 00:17:46.677
tokens on the other end. And how do we

00:17:46.760 --> 00:17:48.997
think about the systems optimization end

00:17:49.080 --> 00:17:51.357
to end? It starts with the data center

00:17:51.440 --> 00:17:53.757
design itself, right? The core compute,

00:17:53.840 --> 00:17:58.557
storage, network, all the accelerators
that go into accelerating each of those

00:17:58.640 --> 00:18:00.197
components.

00:18:00.280 --> 00:18:02.997
How do you even think about the DC to DC

00:18:03.080 --> 00:18:05.637
connectivity and the networking,

00:18:05.720 --> 00:18:10.397
as well as the offload to something
like the local compute, right? That's the

00:18:10.480 --> 00:18:14.597
sort of systems challenge.
But before we even get into

00:18:14.680 --> 00:18:18.797
all of the systems and the
technology and the innovation,

00:18:18.880 --> 00:18:20.357
perhaps the most important design

00:18:20.440 --> 00:18:23.597
criteria for us is how do we earn the

00:18:23.680 --> 00:18:26.117
permission from the communities in which

00:18:26.200 --> 00:18:28.437
we're building these data centers. And

00:18:28.520 --> 00:18:31.597
that's where these principles ground us

00:18:31.680 --> 00:18:33.397
and focus us.

00:18:33.480 --> 00:18:35.957
How do we ensure that the DCs do not

00:18:36.040 --> 00:18:39.397
increase the electricity prices,

00:18:39.480 --> 00:18:44.517
making sure that we are replenishing
all our water use, creating jobs in the

00:18:44.600 --> 00:18:47.517
local communities for the local residents,

00:18:47.600 --> 00:18:51.837
adding to the tax base, making sure
we're strengthening the communities by

00:18:51.920 --> 00:18:54.237
investing in local training and the

00:18:54.320 --> 00:18:57.517
nonprofits in the area. Only when we

00:18:57.600 --> 00:19:01.717
live up to these principles, do the hard
work around it, is when we earn the

00:19:01.800 --> 00:19:05.037
permission to go ahead
and innovate and build.

00:19:05.120 --> 00:19:07.797
And we've been doing a lot of data

00:19:07.880 --> 00:19:11.477
center build out. Today, Uh, Azure spans

00:19:11.560 --> 00:19:16.557
more than 500 data centers in 80 regions.
It's the most expansive, you

00:19:16.640 --> 00:19:17.877
know, we have the most expansive

00:19:17.960 --> 00:19:20.477
hyperscale footprint out there. Um, and

00:19:20.560 --> 00:19:23.637
we've added more data center capacity in

00:19:23.720 --> 00:19:27.437
the last 18 months than the first decade

00:19:27.520 --> 00:19:31.877
of Azure. Just to put that in perspective.
Uh, but more importantly,

00:19:31.960 --> 00:19:36.117
what we're building is also very
different. Uh, in fact, the first 15

00:19:36.200 --> 00:19:38.717
years or so, you know, we built out the

00:19:38.800 --> 00:19:43.237
commercial cloud infrastructure
for a set of heterogeneous,

00:19:43.320 --> 00:19:48.237
you know, workloads that spanned the
enterprise. Uh, but now when you look at

00:19:48.320 --> 00:19:51.317
what we're building, right? When you
think about all the gigawatts that were

00:19:51.400 --> 00:19:55.477
going to come online,
really they have three dominant

00:19:55.560 --> 00:20:00.197
workloads. There is training,
there's inference, and then there is the

00:20:00.280 --> 00:20:04.117
agent runtime. Right, these are three
dominant workloads. And in fact, when

00:20:04.200 --> 00:20:09.437
you look at Fairwater, uh,
it was sort of our first AI super

00:20:09.520 --> 00:20:13.117
factory. You know,
it spanned two regions, uh,

00:20:13.200 --> 00:20:17.877
Georgia and Wisconsin. Uh,
the entire system was designed from

00:20:17.960 --> 00:20:21.557
the ground up for AI. Right, we worked
in fact very closely with even Nvidia on

00:20:21.640 --> 00:20:24.437
this, and it's a two-story architecture

00:20:24.520 --> 00:20:27.397
that lets us essentially place racks,

00:20:27.480 --> 00:20:29.357
obviously in three dimensions, and pack

00:20:29.440 --> 00:20:32.837
the the maximum number of GPUs densely

00:20:32.920 --> 00:20:34.797
uh, with network access. Uh, and that

00:20:34.880 --> 00:20:37.277
means you really got fantastic higher

00:20:37.360 --> 00:20:41.957
performance networking, lower latency,
uh, and more effective bandwidth across

00:20:42.040 --> 00:20:46.397
the entire cluster. Um,
and we're rethinking even the power

00:20:46.480 --> 00:20:48.197
delivery, right? So, we sort of okay,

00:20:48.280 --> 00:20:51.197
how do we deliver hundreds of kilowatts

00:20:51.280 --> 00:20:54.157
per row while minimizing all the loss,

00:20:54.240 --> 00:20:58.277
right? Which is the conversion loss
that happens from the grid to the silicon.

00:20:58.360 --> 00:21:01.117
So, we basically sort of even, uh, took,

00:21:01.200 --> 00:21:03.437
uh, a new approach to it. And all of

00:21:03.520 --> 00:21:06.437
this also changes with, uh, the cooling

00:21:06.520 --> 00:21:10.837
system, right? And water. So,
in fact, the cooling loop is filled

00:21:10.920 --> 00:21:12.837
um once in the data center can operate

00:21:12.920 --> 00:21:15.397
effectively with zero water consumption.

00:21:15.480 --> 00:21:18.517
In fact, the daily water usage over the

00:21:18.600 --> 00:21:21.317
course of an entire year is roughly

00:21:21.400 --> 00:21:23.677
equivalent to what a single restaurant

00:21:23.760 --> 00:21:28.450
would use, right? I mean,
that's um >> [applause]

00:21:30.880 --> 00:21:33.157
>> And when it comes
to the systems and the

00:21:33.240 --> 00:21:35.477
silicon, uh again, we have a lot of

00:21:35.560 --> 00:21:38.037
choice. Uh we have first-party silicon.

00:21:38.120 --> 00:21:40.637
We have partner sys- uh systems.

00:21:40.720 --> 00:21:43.237
Uh we were the very first cloud to bring

00:21:43.320 --> 00:21:48.117
up, in fact, NVIDIA's uh Vera Rubin
system for validation. Very exciting to

00:21:48.200 --> 00:21:52.917
see that. Uh we're working closely with
AMD. We worked with them on MI300. Now,

00:21:53.000 --> 00:21:53.997
we're working with them on their

00:21:54.080 --> 00:21:57.957
next-generation uh AMD GPUs. Uh

00:21:58.040 --> 00:22:02.357
with Maya 200, uh it's continuing
to scale. In fact, it's live in Iowa and

00:22:02.440 --> 00:22:06.397
Arizona. Uh we'll deploy it
internationally later this year. It

00:22:06.480 --> 00:22:09.277
delivers 30% improved tokens per dollar

00:22:09.360 --> 00:22:11.357
compared to sort of what's the leading

00:22:11.440 --> 00:22:14.837
uh GPU today. Uh and we have validated

00:22:14.920 --> 00:22:17.637
it with 55 GPT-55, and we're going to

00:22:17.720 --> 00:22:20.277
use that to power uh Microsoft 365

00:22:20.360 --> 00:22:23.677
Copilot. Uh and so, when it comes to

00:22:23.760 --> 00:22:25.917
running these um

00:22:26.000 --> 00:22:28.797
agents, the interesting thing now is

00:22:28.880 --> 00:22:30.797
it's no longer just about having an AI

00:22:30.880 --> 00:22:34.997
accelerator or a GPU. The CPU is really

00:22:35.080 --> 00:22:36.757
critical. In fact, the ratios may be

00:22:36.840 --> 00:22:39.837
even coming to 1:1. Uh and that's why

00:22:39.920 --> 00:22:42.517
we're innovating uh with Cobalt. Uh

00:22:42.600 --> 00:22:45.157
we're announcing the preview of Cobalt

00:22:45.240 --> 00:22:48.517
200 VMs, our next-generation Arm-based

00:22:48.600 --> 00:22:51.717
CPU design for both cloud-native and

00:22:51.800 --> 00:22:57.117
agent uh workloads uh today. So, it's
exciting to see uh Cobalt make progress

00:22:57.200 --> 00:23:00.810
as well. >> [applause]

00:23:02.040 --> 00:23:05.757
>> You know, one thing that uh we've
been trying to make sure is that being

00:23:05.840 --> 00:23:08.997
optimized for these new workloads I
talked about, right? These new Asian

00:23:09.080 --> 00:23:13.037
workloads. So, in fact, Cobalt delivers
50 plus percent better performance than

00:23:13.120 --> 00:23:15.357
Cobalt 100 on cloud native. But, we

00:23:15.440 --> 00:23:20.157
started benchmarking them using the
GitHub Copilot traces, uh these agentic

00:23:20.240 --> 00:23:23.757
traces to see because the call patterns
are so different. Uh and we're now

00:23:23.840 --> 00:23:26.917
seeing 33% lower latency for the agent

00:23:27.000 --> 00:23:30.917
calls, uh 14% faster speed, 23% higher

00:23:31.000 --> 00:23:33.117
throughput. So, this is about the core

00:23:33.200 --> 00:23:36.197
design of both the AI accelerator and

00:23:36.280 --> 00:23:40.157
the CPU uh for the agent. And of course,

00:23:40.240 --> 00:23:44.517
um when you talk about AI workloads,
you need scale, you need reliability, and

00:23:44.600 --> 00:23:46.917
that's why the network becomes super

00:23:47.000 --> 00:23:49.717
critical. Uh we have innovated with the

00:23:49.800 --> 00:23:51.917
MRC architecture

00:23:52.000 --> 00:23:54.797
uh and rebuilt how traffic across Azure

00:23:54.880 --> 00:23:57.757
moves to support AI workloads, uh which

00:23:57.840 --> 00:24:02.157
are fundamentally synchronous data
parallel uh workloads, right? And so,

00:24:02.240 --> 00:24:06.237
therefore, they span tens
and thousands of GPUs.

00:24:06.320 --> 00:24:10.157
Um and so, you need to have them coherent.
That's what the next frontier

00:24:10.240 --> 00:24:13.237
of you know, is for us to make sure uh

00:24:13.320 --> 00:24:17.477
we're able to keep scaling uh
the network architecture. And of course,

00:24:17.560 --> 00:24:20.917
it's not about just inside the data
center, it's also about connecting

00:24:21.000 --> 00:24:23.677
across the data centers. Um and every

00:24:23.760 --> 00:24:26.517
Fairwater data center, for example, is

00:24:26.600 --> 00:24:29.317
connected through our continent-spanning

00:24:29.400 --> 00:24:32.637
uh AI WAN, forming this truly uh

00:24:32.720 --> 00:24:37.677
fungible uh compute fabric.
Um and you know,

00:24:37.760 --> 00:24:39.277
you know, all this innovation is

00:24:39.360 --> 00:24:44.957
exciting, but when you think about
innovation, designing for AI, and I'll

00:24:45.040 --> 00:24:49.677
say deep understanding of systems,

00:24:49.760 --> 00:24:51.757
um it you know, there's no better

00:24:51.840 --> 00:24:53.997
company than NVIDIA, and there's no

00:24:54.080 --> 00:24:57.237
better person uh than Jensen uh to talk

00:24:57.320 --> 00:24:59.837
about it. So, I wanted to invite live

00:24:59.920 --> 00:25:03.357
from Taipei, uh Jensen Huang, CEO and uh

00:25:03.440 --> 00:25:07.880
founder of Nvidia.
>> [applause and cheering]

00:25:11.040 --> 00:25:14.197
>> Thank you. >> Jensen,
thank you so much for being at

00:25:14.280 --> 00:25:19.317
Build again. I know it's late for you
in Taipei. I really appreciate you staying

00:25:19.400 --> 00:25:21.717
up. Uh you know,

00:25:21.800 --> 00:25:26.637
and then, you know, I'm I've been
looking at social and, you know, and

00:25:26.720 --> 00:25:29.757
everything people have been talking
about since your keynote over the

00:25:29.840 --> 00:25:33.237
weekend. Uh and suddenly, you know, this

00:25:33.320 --> 00:25:36.837
concept of unmetered intelligence right

00:25:36.920 --> 00:25:41.677
at the edge is so hot again. So, maybe
you want to talk a little bit. You've

00:25:41.760 --> 00:25:46.117
thought about this, talked about this,
and now, of course, with RTX Spark

00:25:46.200 --> 00:25:48.197
really delivered, I think, what's a

00:25:48.280 --> 00:25:51.717
breakthrough uh system for AI to be much

00:25:51.800 --> 00:25:53.317
more ubiquitous. But maybe, Jensen, you

00:25:53.400 --> 00:25:55.677
can just share a little bit your vision

00:25:55.760 --> 00:25:59.117
around where you see this going.

00:25:59.200 --> 00:26:02.877
>> Well, this all started about 3 years
ago between a conversation between you and

00:26:02.960 --> 00:26:05.717
I. And we were talking about how

00:26:05.800 --> 00:26:08.877
we could build a new class of PCs that's

00:26:08.960 --> 00:26:11.637
incredible for designers and creators.

00:26:11.720 --> 00:26:15.597
And uh it would be incredible for
artificial intelligence. And it would be

00:26:15.680 --> 00:26:17.917
one of these systems that uh

00:26:18.000 --> 00:26:22.677
has the processing capability, but also
the software stack that's integrated

00:26:22.760 --> 00:26:26.557
into the world's design packages and
creator packages. And And of course, all

00:26:26.640 --> 00:26:30.397
the things that we're doing
with AI. And here we are uh

00:26:30.480 --> 00:26:32.997
3 years later, we built an incredible

00:26:33.080 --> 00:26:34.237
new chip.

00:26:34.320 --> 00:26:38.677
And this system is supported by uh all

00:26:38.760 --> 00:26:40.237
of this new software that you create for

00:26:40.320 --> 00:26:43.117
Windows. And we now have the ability to

00:26:43.200 --> 00:26:48.477
have essentially an autonomous
agent running on the PC.

00:26:48.560 --> 00:26:51.877
Now, when you take a step back,
and you think about, what does that mean?

00:26:51.960 --> 00:26:56.237
For For the 40 years that or some 30
years we've been working together, we

00:26:56.320 --> 00:26:57.597
went from

00:26:57.680 --> 00:27:01.037
inventing DirectX together to

00:27:01.120 --> 00:27:04.037
creating now this

00:27:04.120 --> 00:27:06.557
incredible computer that has autonomous

00:27:06.640 --> 00:27:10.437
systems running. The PC evolved from

00:27:10.520 --> 00:27:13.677
being an incredible tool to now being a

00:27:13.760 --> 00:27:16.477
tool that's used autonomously by an AI

00:27:16.560 --> 00:27:19.517
assistant. And so, the idea that I could

00:27:19.600 --> 00:27:22.477
be traveling and I'm on the phone and I

00:27:22.560 --> 00:27:25.957
could text my PC and ask my PC to get

00:27:26.040 --> 00:27:28.437
some coding done or some idea that I

00:27:28.520 --> 00:27:33.077
have and it would fire
up the tools on the PC.

00:27:33.160 --> 00:27:37.237
And it would make the modifications
or the changes or the design that I was I

00:27:37.320 --> 00:27:39.917
told it to to do and it would iterate

00:27:40.000 --> 00:27:45.277
with me while I'm away from the PC.
My PC became an assistant. While I'm

00:27:45.360 --> 00:27:46.917
sitting there, of course, that this PC

00:27:47.000 --> 00:27:48.837
would be my great assistant as well. And

00:27:48.920 --> 00:27:52.597
so, this idea that that the PC evolved

00:27:52.680 --> 00:27:54.837
from a from a personal computer to a

00:27:54.920 --> 00:27:58.277
personal AI is just really exciting and

00:27:58.360 --> 00:28:03.397
to see it come to life, Satya, to see
it come to life and actually doing that.

00:28:03.480 --> 00:28:06.837
You know, so I'm I'm super
excited about it. I Spark

00:28:06.920 --> 00:28:10.757
you you mentioned earlier has all this
incredible capabilities, a petaflop of

00:28:10.840 --> 00:28:13.717
AI performance. It has a petaflops of

00:28:13.800 --> 00:28:17.997
NVFP4, this numerical format that our two

00:28:18.080 --> 00:28:21.517
companies worked
on together that allows us to

00:28:21.600 --> 00:28:24.517
take advantage of this 128 GB of memory

00:28:24.600 --> 00:28:29.477
and fit maybe a couple of hundred
billion parameter model. A couple of

00:28:29.560 --> 00:28:33.757
hundred billion parameter model
is state-of-the-art. And so, I think the

00:28:33.840 --> 00:28:37.997
days of having a really smart assistant
running autonomously on the PC is here.

00:28:38.080 --> 00:28:41.277
>> Yeah, I know it's so awesome and
in fact, I'm also excited about Windows

00:28:41.360 --> 00:28:45.517
coming to the GB300. And so,
that's another thing that it's

00:28:45.600 --> 00:28:49.397
kind of like data center right
on your desktop and it's so exciting. But

00:28:49.480 --> 00:28:53.397
talking about that um data
center side, obviously,

00:28:53.480 --> 00:28:56.157
you know, this entire thing got started

00:28:56.240 --> 00:28:58.997
when we built the first supercomputer uh

00:28:59.080 --> 00:29:02.037
together to train uh the GPT models and

00:29:02.120 --> 00:29:04.357
we've come a long way. In fact, even I

00:29:04.440 --> 00:29:06.357
was talking about the Fairwater design,

00:29:06.440 --> 00:29:09.277
it is custom-built essentially for uh

00:29:09.360 --> 00:29:11.597
the Grace Blackwell era uh to be able to

00:29:11.680 --> 00:29:16.917
max the the data center design with
the system design you had. And now, of

00:29:17.000 --> 00:29:20.117
course, we're validating Vera Rubin,
we're very excited about it. Maybe you

00:29:20.200 --> 00:29:24.517
want to share a little bit about sort
of what happens even on the cloud side uh

00:29:24.600 --> 00:29:28.517
with how you're pushing
on the systems innovation.

00:29:28.600 --> 00:29:30.757
>> Well, our journey has been incredible.

00:29:30.840 --> 00:29:35.477
We built the first AI supercomputer
together. That was based on Ampere.

00:29:35.560 --> 00:29:39.197
Of course, Hopper was an incredible
success. This first two generations were

00:29:39.280 --> 00:29:41.757
focused on pre-training. Grace Blackwell

00:29:41.840 --> 00:29:44.317
came along and

00:29:44.400 --> 00:29:47.637
all of the focus moved to post-training,

00:29:47.720 --> 00:29:52.197
uh reinforcement learning, which allowed
us to have reasoning models, and these

00:29:52.280 --> 00:29:56.477
reasoning models based on mixture
of experts were incredibly intelligent,

00:29:56.560 --> 00:29:58.077
energy-efficient,

00:29:58.160 --> 00:30:01.197
um but it requires uh giant systems. And

00:30:01.280 --> 00:30:04.677
so, we created NVLink 72, and the entire

00:30:04.760 --> 00:30:07.997
rack became one computer. We had evolved

00:30:08.080 --> 00:30:12.797
from one node to now one rack. Well,

00:30:12.880 --> 00:30:16.837
Microsoft deployed
the largest number of Grace

00:30:16.920 --> 00:30:18.717
Blackwells in the world today.

00:30:18.800 --> 00:30:22.277
The fastest and the largest number
of Grace Blackwells in the world.

00:30:22.360 --> 00:30:26.397
Fairwater is just
a magnificent system to look at.

00:30:26.480 --> 00:30:31.237
It's just a miracle of engineering.
It's It's just an incredible feat. Uh it's

00:30:31.320 --> 00:30:34.877
completely liquid-cooled. You mentioned
something earlier that I'm very proud of

00:30:34.960 --> 00:30:37.357
as well, that it's close-looped.

00:30:37.440 --> 00:30:39.957
Basically, uses almost no water, and

00:30:40.040 --> 00:30:42.477
it's incredibly environmentally uh

00:30:42.560 --> 00:30:45.037
it's energy efficient. We're able to

00:30:45.120 --> 00:30:47.717
increase the token generation rate and

00:30:47.800 --> 00:30:50.517
reduce the cost of token generation by

00:30:50.600 --> 00:30:53.397
an order of magnitude, some 30 times

00:30:53.480 --> 00:30:57.997
over Hopper. So, that was a huge
achievement. Well, Vera Rubin was

00:30:58.080 --> 00:31:01.077
created for a world where

00:31:01.160 --> 00:31:05.357
these AIs are now agentic.
And so, whereas Hopper was created for

00:31:05.440 --> 00:31:09.197
pre-training, Grace Blackwell
for training, post-training, and also

00:31:09.280 --> 00:31:10.557
inference,

00:31:10.640 --> 00:31:14.037
Vera Rubin is designed to run agents.

00:31:14.120 --> 00:31:16.717
It's agents, as you know, this computing

00:31:16.800 --> 00:31:18.837
pattern is exactly the same computing

00:31:18.920 --> 00:31:22.397
pattern when a run on the RTX Spark.

00:31:22.480 --> 00:31:26.197
It's exactly the same agentic system,
except of course, it's going to be much,

00:31:26.280 --> 00:31:31.477
much larger. We'll process enormous
number of them simultaneously. Many of

00:31:31.560 --> 00:31:34.677
them are going to be from different
customers and different different

00:31:34.760 --> 00:31:37.437
partners. And so, the entire path, the

00:31:37.520 --> 00:31:39.757
entire coding path from

00:31:39.840 --> 00:31:45.437
storage, which is the long-term memory,
the working memory, is encrypted.

00:31:45.520 --> 00:31:48.917
The data is encrypted in transit, the data

00:31:49.000 --> 00:31:51.637
is also encrypted in use. And so, we've

00:31:51.720 --> 00:31:55.997
we're going to really innovate in the
area of confidential computing. And so,

00:31:56.080 --> 00:31:58.397
this entire disaggregated, distributed

00:31:58.480 --> 00:32:00.797
computing system, you mentioned CPUs,

00:32:00.880 --> 00:32:05.677
Vera is a revolutionary CPU designed
for agents. You know, the past CPUs were

00:32:05.760 --> 00:32:08.917
designed for humans.
And you know, we're just more patient

00:32:09.000 --> 00:32:14.437
than than agents are. And agents uh
want low latency, just as have you

00:32:14.520 --> 00:32:18.077
been working on as well. Vera was
designed for extremely low latency. And

00:32:18.160 --> 00:32:22.477
so, Vera Rubin's is just completely
revolutionary. I can't wait uh to show

00:32:22.560 --> 00:32:24.637
it to everybody.
You've already stood it up.

00:32:24.720 --> 00:32:27.277
>> Yep. >> Our two teams
Our two teams have been

00:32:27.360 --> 00:32:28.997
working very closely. And you know,

00:32:29.080 --> 00:32:32.597
almost uh long before the chips taped

00:32:32.680 --> 00:32:36.597
out, long before the systems were
brought up, our two teams were already

00:32:36.680 --> 00:32:40.877
completely aligned. And so, the design
the data centers were created for Vera

00:32:40.960 --> 00:32:45.037
Rubin. Vera Rubin is designed
and integrated into your complete stack,

00:32:45.120 --> 00:32:49.117
into your networking, into your security.
And so, the moment that our

00:32:49.200 --> 00:32:53.317
systems were rolling off the lines,
they were being stood up at Microsoft. So,

00:32:53.400 --> 00:32:55.557
I'm incredibly excited
about the collaboration.

00:32:55.640 --> 00:32:58.997
>> Yeah, I know this speed of light
execution between the teams is fantastic

00:32:59.080 --> 00:33:01.477
to see. And of course, all this to is to

00:33:01.560 --> 00:33:06.197
power the ecosystem around us, right?
I mean, that you and I, having grown up

00:33:06.280 --> 00:33:10.957
with the PC, the server, and now with AI,
have always thought about ultimately

00:33:11.040 --> 00:33:15.077
it's about creating the opportunity
for every developer, every organization to

00:33:15.160 --> 00:33:18.637
build on the work that we do and the
platforms we create. And speaking of

00:33:18.720 --> 00:33:21.157
that, there's a lot of software uh that

00:33:21.240 --> 00:33:25.757
NVIDIA builds that's all also coming.
Uh for example, we're going to have your

00:33:25.840 --> 00:33:29.597
models in Foundry, your tooling
in Foundry. We're going to have, in fact,

00:33:29.680 --> 00:33:34.197
uh your software even help us with
accelerating our uh workloads when it

00:33:34.280 --> 00:33:37.917
comes to even the data warehouse.
We're going to obviously have uh stuff in

00:33:38.000 --> 00:33:40.157
Windows. Just talk a little bit about

00:33:40.240 --> 00:33:45.277
that broader vision of what does it mean
for us an opportunity, right? Because

00:33:45.360 --> 00:33:46.917
everybody talks about this one model or

00:33:47.000 --> 00:33:49.037
one piece of tech. But it's about the

00:33:49.120 --> 00:33:52.797
broadest, biggest opportunity for people

00:33:52.880 --> 00:33:57.757
to create value. Uh maybe you want
to share a little bit about that.

00:33:57.840 --> 00:34:00.597
>> Well, we've been preparing for this
moment. You know, what happened in the

00:34:00.680 --> 00:34:05.237
last several months, we've been working
for a decade and a half together.

00:34:05.320 --> 00:34:08.557
Getting ready for really what happened
in the last several months. All of a

00:34:08.640 --> 00:34:11.197
sudden, because of agentic systems, the

00:34:11.280 --> 00:34:12.757
convergence of these really great

00:34:12.840 --> 00:34:18.517
models, AI is now useful.
If you just look at GitHub,

00:34:18.600 --> 00:34:22.717
the the the the commits
into GitHub has gone

00:34:22.800 --> 00:34:24.957
completely parabolic. In the last

00:34:25.040 --> 00:34:27.197
several months, the the number of

00:34:27.280 --> 00:34:30.037
commits increased by factor of three.

00:34:30.120 --> 00:34:35.037
It's clear that agentic systems are
useful, that it's doing productive work,

00:34:35.120 --> 00:34:36.597
and also

00:34:36.680 --> 00:34:39.357
tokens are now profitable as a result.

00:34:39.440 --> 00:34:42.237
And so, the amount of demand for compute

00:34:42.320 --> 00:34:45.437
between the the the usage of the AI and

00:34:45.520 --> 00:34:49.557
the computation that's necessary
for agents, the compute demand has really

00:34:49.640 --> 00:34:53.517
gone through the roof. Well, one of
the things that we've been we been doing

00:34:53.600 --> 00:34:57.917
together is making sure that all of the
tools that the agents are going to use

00:34:58.000 --> 00:35:00.077
are fully accelerated. Fabric, for

00:35:00.160 --> 00:35:02.917
example, is now fully accelerated. We're

00:35:03.000 --> 00:35:06.917
accelerating data processing, um SQL, uh

00:35:07.000 --> 00:35:11.557
Spark, uh semantic-based, vector-based,

00:35:11.640 --> 00:35:15.117
uh graph-based. We're going to make sure
that all of the tools that are available

00:35:15.200 --> 00:35:19.237
on Azure are going to be fully GPU
accelerated because the agents are going

00:35:19.320 --> 00:35:23.237
to be impatient. The faster we can
get the answers back to the agents, the

00:35:23.320 --> 00:35:26.877
faster they can iterate, the faster
they can generate tokens, which are

00:35:26.960 --> 00:35:30.637
ultimately what the developers,
both of our customers, would like to do is

00:35:30.720 --> 00:35:34.797
generate a lot of tokens that are really
profitable, that are highly intelligent.

00:35:34.880 --> 00:35:39.077
>> Now, thank you so much, Jensen, for
the partnership and the leadership and the

00:35:39.160 --> 00:35:41.477
innovation uh that you bring to this

00:35:41.560 --> 00:35:46.597
entire ecosystem, and really thrilled to
be working closely with you and the team

00:35:46.680 --> 00:35:50.797
and bring all this to the developers
here and beyond, and look forward to

00:35:50.880 --> 00:35:55.357
seeing what the next sort of few months
and the next year bring in terms of the

00:35:55.440 --> 00:35:59.357
innovation that gets built on top
of the platform. So, thank you again for

00:35:59.440 --> 00:36:02.757
joining this late
in the night uh from Taipei.

00:36:02.840 --> 00:36:07.470
>> Thank you so much, Satya, for your
partnership and friendship. [applause]

00:36:11.520 --> 00:36:16.437
>> So, so far, we've talked
about the edge and the cloud.

00:36:16.520 --> 00:36:19.077
Um the current form
factors, right? I mean,

00:36:19.160 --> 00:36:23.237
when I saw that Jensen picture from the
weekend where he had all the desktops, I

00:36:23.320 --> 00:36:25.077
felt like, "Man, I'm back in the '90s,"

00:36:25.160 --> 00:36:28.037
right? Because it was so cool to see the

00:36:28.120 --> 00:36:31.357
lineup of all the machines that I loved

00:36:31.440 --> 00:36:34.157
and I grew up with back yet again with

00:36:34.240 --> 00:36:36.597
new functionality, right? The same form

00:36:36.680 --> 00:36:39.277
factor, but unbelievable new

00:36:39.360 --> 00:36:43.797
functionality because of the onboard
AI capability, right? So, that's sort of

00:36:43.880 --> 00:36:47.597
what we've seen with the laptop,
the desktop, and of

00:36:47.680 --> 00:36:52.317
course with the cloud. But,
it also, you know, sets up that

00:36:52.400 --> 00:36:57.237
next question. If you have
that capability, which is

00:36:57.320 --> 00:36:59.197
new function, and you can put it into

00:36:59.280 --> 00:37:01.397
existing form factors,

00:37:01.480 --> 00:37:04.557
can you even purpose-build

00:37:04.640 --> 00:37:06.837
new form factors

00:37:06.920 --> 00:37:08.677
for the new function? Can you build a

00:37:08.760 --> 00:37:11.597
new platform even for

00:37:11.680 --> 00:37:14.237
uh the agent era? And that is the

00:37:14.320 --> 00:37:17.597
motivation behind Project Solara, which

00:37:17.680 --> 00:37:19.717
we're introducing today. And to talk

00:37:19.800 --> 00:37:21.957
about this, I wanted to invite Stevie on

00:37:22.040 --> 00:37:25.037
stage, but first, let's roll the video.

00:37:25.120 --> 00:37:28.557
>> First, there's you.

00:37:28.640 --> 00:37:32.597
Then, everything in front of you.

00:37:32.680 --> 00:37:39.120
We get it. The noise,
the weight of it all.

00:37:39.880 --> 00:37:43.157
And so, we've [music]
been building something.

00:37:43.240 --> 00:37:45.557
A new chip-to-cloud platform for an

00:37:45.640 --> 00:37:48.277
agent-first world, and for new

00:37:48.360 --> 00:37:53.237
agent-first devices. You come close,

00:37:53.320 --> 00:37:57.037
it's ready when you are.
>> [music] >> You speak,

00:37:57.120 --> 00:38:01.277
it understands you.
What matters comes forward.

00:38:01.360 --> 00:38:05.437
Everything else [music] falls away.

00:38:05.520 --> 00:38:10.357
And when you want it to keep
going, it keeps going.

00:38:10.440 --> 00:38:12.317
Your call. >> [music]

00:38:12.400 --> 00:38:14.797
>> Sometimes it sits with you.

00:38:14.880 --> 00:38:19.277
Sometimes it goes with you.
Sometimes it [music] sees what you want

00:38:19.360 --> 00:38:21.437
it to see.

00:38:21.520 --> 00:38:26.317
It doesn't decide for you.
It just lights the path

00:38:26.400 --> 00:38:30.037
and tries to clear [music]
the way when you need it to.

00:38:30.120 --> 00:38:35.477
That's the idea.
A whole ecosystem built to clear the way

00:38:35.560 --> 00:38:38.877
so all that's left is [music]
where you need to go.

00:38:38.960 --> 00:38:42.577
Wherever you are, whenever it matters.

00:38:42.660 --> 00:38:44.680
>> [music]

00:38:47.520 --> 00:38:51.437
>> I am so excited to be here.

00:38:51.520 --> 00:38:54.517
It really is great
to be back on this stage.

00:38:54.600 --> 00:38:58.637
Now, before I talk about those
awesome new devices you just saw,

00:38:58.720 --> 00:39:01.277
let me start with the why.

00:39:01.360 --> 00:39:04.477
Back at Build 2023, I talked about the

00:39:04.560 --> 00:39:07.397
outside AI application structure.

00:39:07.480 --> 00:39:09.957
Where AI moved from operating within the

00:39:10.040 --> 00:39:12.957
application frame to operating globally,

00:39:13.040 --> 00:39:14.997
working across multiple apps and

00:39:15.080 --> 00:39:20.117
services to connect, coordinate,
and maintain context across entire

00:39:20.200 --> 00:39:21.437
workflows,

00:39:21.520 --> 00:39:25.077
devices, and time scales.

00:39:25.160 --> 00:39:27.397
What if there were an ecosystem of

00:39:27.480 --> 00:39:30.197
devices specifically designed for that

00:39:30.280 --> 00:39:32.437
new type of application structure, for

00:39:32.520 --> 00:39:35.877
those types of agents, for that

00:39:35.960 --> 00:39:38.957
transformational interaction technology?

00:39:39.040 --> 00:39:42.477
That is the impetus behind Project Solara.

00:39:42.560 --> 00:39:44.957
But with so many possible forms, which

00:39:45.040 --> 00:39:50.037
one do you pick? What is the next device?

00:39:50.120 --> 00:39:52.277
>> [snorts] >> You see,
the big aha for us is that it's

00:39:52.360 --> 00:39:56.637
not about choosing one specific form
factor. It is about creating a system

00:39:56.720 --> 00:39:58.197
that extends your agent across a

00:39:58.280 --> 00:40:01.517
constellation of devices.

00:40:01.600 --> 00:40:04.037
The next computer is not one device.

00:40:04.120 --> 00:40:06.997
It is all these devices working together

00:40:07.080 --> 00:40:09.837
as one system with agents showing up

00:40:09.920 --> 00:40:14.357
closer to where and when you need them.

00:40:14.440 --> 00:40:16.317
To really To realize this vision, two

00:40:16.400 --> 00:40:20.197
challenges immediately show up. First,

00:40:20.280 --> 00:40:22.117
many specialized form factors already

00:40:22.200 --> 00:40:25.437
exist, but often rely on custom one-off

00:40:25.520 --> 00:40:29.517
apps and fragmented stacks that are
difficult and expensive to build,

00:40:29.600 --> 00:40:31.797
deploy, and maintain.

00:40:31.880 --> 00:40:34.597
And second, across every industry,

00:40:34.680 --> 00:40:39.077
people and organizations already
building their own agents, deeply

00:40:39.160 --> 00:40:41.317
specialized and instrumented for their

00:40:41.400 --> 00:40:46.397
work. But the impact of those agents
is constrained by how and where they can

00:40:46.480 --> 00:40:50.117
exist. Project Volterra addresses both by

00:40:50.200 --> 00:40:52.117
giving organizations a way to extend

00:40:52.200 --> 00:40:55.237
their agents onto new, purpose-built,

00:40:55.320 --> 00:41:00.077
easy-to-manage form factors designed
to reach the nooks and crannies where

00:41:00.160 --> 00:41:03.877
conventional computers
either do not exist or

00:41:03.960 --> 00:41:06.677
are not optimal.

00:41:06.760 --> 00:41:11.437
It's a turnkey solution for building
unique agent-first devices enabled by

00:41:11.520 --> 00:41:13.517
three pillars.

00:41:13.600 --> 00:41:16.637
First, it's enterprise-ready, enabled by

00:41:16.720 --> 00:41:21.397
the AOSP-based Microsoft
device ecosystem platform.

00:41:21.480 --> 00:41:25.597
Second, it's it has
an agent-driven interaction

00:41:25.680 --> 00:41:28.357
model with just-in-time UI that adapts

00:41:28.440 --> 00:41:30.997
to the form factor. And third, it has

00:41:31.080 --> 00:41:33.557
extensibility, so you can bring your own

00:41:33.640 --> 00:41:35.837
agents. And tying it all together is

00:41:35.920 --> 00:41:40.917
Azure, unifying the system
across cloud and device. Okay.

00:41:41.000 --> 00:41:44.637
That's enough of that. Now,
let's talk about the devices.

00:41:44.720 --> 00:41:48.637
Today, we're previewing previewing
two very broad categories. The first is

00:41:48.720 --> 00:41:51.397
stationary, and the second is portable.

00:41:51.480 --> 00:41:55.477
The first is a device is designed
for your desk, and it's built on MediaTek

00:41:55.560 --> 00:41:58.957
silicon. With Hello
for Business, just walking up

00:41:59.040 --> 00:42:01.357
to the device securely signs you in,

00:42:01.440 --> 00:42:04.157
giving you direct access to your agents,

00:42:04.240 --> 00:42:08.557
just like Nathan's about to show you here.

00:42:08.640 --> 00:42:10.477
For the information worker, this means

00:42:10.560 --> 00:42:12.597
frictionless yet protected access to

00:42:12.680 --> 00:42:15.757
Microsoft 365 Copilot grounded in Work

00:42:15.840 --> 00:42:19.757
IQ. And with a simple glance,
it surfaces what matters next in your

00:42:19.840 --> 00:42:22.237
workday, helping you think, plan, and

00:42:22.320 --> 00:42:24.397
even act by delegating tasks to your

00:42:24.480 --> 00:42:27.997
agents with a simple tap or just using

00:42:28.080 --> 00:42:31.557
your voice. Think of it
as a dedicated secured

00:42:31.640 --> 00:42:35.837
ambient device for work.
It even supports experiences like

00:42:35.920 --> 00:42:37.917
handoff between devices, acting as a

00:42:38.000 --> 00:42:41.397
companion to your existing Windows PC.

00:42:41.480 --> 00:42:44.317
Or it can even let you access your cloud

00:42:44.400 --> 00:42:50.520
PC through Windows 365 and a connected
monitor. How cool is that?

00:42:50.900 --> 00:42:52.920
>> [applause]

00:42:54.560 --> 00:42:57.157
>> Now, the second device is portable.

00:42:57.240 --> 00:43:02.077
It's reimagining a wearable that
millions of people use every day,

00:43:02.160 --> 00:43:05.797
the access badge.
Built using Qualcomm silicon for

00:43:05.880 --> 00:43:10.037
wearable, this digital badge is a
lightweight form factor designed for

00:43:10.120 --> 00:43:12.797
agent interactions on the go.

00:43:12.880 --> 00:43:17.237
And then it's adaptable across
a variety of verticals and workflows.

00:43:17.320 --> 00:43:22.560
All right, I have here
an early prototype of the badge.

00:43:22.880 --> 00:43:25.037
And so, thanks. And using my

00:43:25.120 --> 00:43:27.917
fingerprint, I tap to unlock the device

00:43:28.000 --> 00:43:32.637
and I have access now to all my agents
in a secured manner. And would you look

00:43:32.720 --> 00:43:36.757
at that, I already have a task. And it
says, "Gather content for your social

00:43:36.840 --> 00:43:41.197
media post for today." So, why not just
do it right now, right? So, I'm going to

00:43:41.280 --> 00:43:44.637
hit record. And then now
the device's camera is

00:43:44.720 --> 00:43:47.957
recording. I'm going to pan across.
I hope you don't mind. I'm going to take

00:43:48.040 --> 00:43:50.917
your shots. Yes, thank you. Copilot,

00:43:51.000 --> 00:43:53.117
find some good shots from this, clean

00:43:53.200 --> 00:43:58.077
them up, and then send them to me so
for the for me and my team to review. All

00:43:58.160 --> 00:44:03.077
right. And then there you have it. Now
my agent's off running through multiple

00:44:03.160 --> 00:44:06.117
tasks to actually clean this up and send
them to me and the team. That's pretty

00:44:06.200 --> 00:44:10.870
cool. >> [applause]

00:44:11.880 --> 00:44:16.397
>> Okay, I know it's a simple demo,
but it's all agent driven and there are so

00:44:16.480 --> 00:44:20.277
many verticals, so many opportunities.
I mean, for example, imagine in health

00:44:20.360 --> 00:44:24.437
care, from the moment you pick up
the device, the right agent shapes the

00:44:24.520 --> 00:44:26.397
experience around the role and the

00:44:26.480 --> 00:44:29.077
workflow, helping with check-ins,

00:44:29.160 --> 00:44:31.837
patient records, and critical insights,

00:44:31.920 --> 00:44:36.557
all through enterprise-grade
secure access. And with the built-in

00:44:36.640 --> 00:44:38.237
microphones, the nurse can start a

00:44:38.320 --> 00:44:40.557
hands-free voice-based documentation,

00:44:40.640 --> 00:44:43.597
including diarization and annotation.

00:44:43.680 --> 00:44:45.757
And the side-facing camera can be used

00:44:45.840 --> 00:44:50.637
to verify and document patient vitals or
even scan in medications and help verify

00:44:50.720 --> 00:44:52.637
workflows. You know, these are just a

00:44:52.720 --> 00:44:57.717
few examples of how this small,
purpose-built wearable can can bring

00:44:57.800 --> 00:44:59.477
intelligence directly into the flow of

00:44:59.560 --> 00:45:02.317
patient care, helping nurses access and

00:45:02.400 --> 00:45:05.677
gather and even act on information while

00:45:05.760 --> 00:45:09.317
staying present with the patient.

00:45:09.400 --> 00:45:12.517
And while both the stationary
and portable concept devices represent a

00:45:12.600 --> 00:45:17.277
specific expression of agent first
devices, their core reference hardware

00:45:17.360 --> 00:45:20.437
and software are designed
to be highly flexible.

00:45:20.520 --> 00:45:24.197
I mean, just with a few changes, loading
a different agent, adjusting the shape,

00:45:24.280 --> 00:45:27.397
the screen size, the sensors,
or even input methods,

00:45:27.480 --> 00:45:30.597
the same foundation, the same software

00:45:30.680 --> 00:45:34.837
can be adapted for many verticals
and workflows, such as retail and

00:45:34.920 --> 00:45:38.517
industrial, hospitality, financial
services, legal, and so forth. I mean,

00:45:38.600 --> 00:45:42.797
that is the power of the platform,
that flexibility.

00:45:42.880 --> 00:45:46.437
So, whatever your scenario, there are
thousands of untapped opportunities to

00:45:46.520 --> 00:45:48.277
bring agent workflows into places where

00:45:48.360 --> 00:45:52.397
computing has not naturally fit before.

00:45:52.480 --> 00:45:54.317
And while this is an early look, we're

00:45:54.400 --> 00:45:56.717
really excited that AccuWeather, Best

00:45:56.800 --> 00:45:59.837
Buy, CVS Health, Levi's, Target, and

00:45:59.920 --> 00:46:02.957
others are working towards exploring how

00:46:03.040 --> 00:46:04.597
specialized agents and devices can

00:46:04.680 --> 00:46:08.197
improve their workflows.

00:46:08.280 --> 00:46:10.677
This is the broader opportunity for the

00:46:10.760 --> 00:46:13.757
ecosystem. Agents moving outside the app

00:46:13.840 --> 00:46:16.757
and taking shape in devices designed for

00:46:16.840 --> 00:46:18.677
a specific scenario, a specific

00:46:18.760 --> 00:46:22.717
customer, and a specific place.

00:46:22.800 --> 00:46:24.477
And for all of you,

00:46:24.560 --> 00:46:27.517
this is the moment to imagine where your

00:46:27.600 --> 00:46:30.117
agents should live, what form they

00:46:30.200 --> 00:46:36.360
should take, and what
new work they can unlock.

00:46:36.600 --> 00:46:40.677
Last week Satya sat down with Cristiano
Amon from Qualcomm to discuss this

00:46:40.760 --> 00:46:44.157
future. Let's roll the video.

00:46:44.240 --> 00:46:46.077
>> Cristiano, it's so so wonderful having

00:46:46.160 --> 00:46:49.117
you here at Device Lab and talk about

00:46:49.200 --> 00:46:51.557
these new reference designs.

00:46:51.640 --> 00:46:56.197
>> Very happy to be here. I have been
to this Device Lab a lot recently. You

00:46:56.280 --> 00:47:00.917
know, one of the things that you and I
have chatted for a while is how there's

00:47:01.000 --> 00:47:02.757
a real platform shift. We are moving

00:47:02.840 --> 00:47:06.237
from building operating systems, devices

00:47:06.320 --> 00:47:11.077
for apps to agents. You want to talk
a little bit about how you see this?

00:47:11.160 --> 00:47:12.917
>> Yeah, absolutely.
In every generation of

00:47:13.000 --> 00:47:14.717
technology transition we've seen a shift

00:47:14.800 --> 00:47:16.917
and I think this one seems to be a very

00:47:17.000 --> 00:47:20.317
big one. Agents really changes the whole

00:47:20.400 --> 00:47:23.597
nature of the device in itself. Starting

00:47:23.680 --> 00:47:28.597
with the fact that, you know,
if the AI understand the world the way we

00:47:28.680 --> 00:47:31.917
understand it, it's going to be closer
to our senses. It's going to be closer

00:47:32.000 --> 00:47:33.597
to our eyes, to our mouth, to our ears.

00:47:33.680 --> 00:47:36.837
It's going to be things
that we wear and it's changing

00:47:36.920 --> 00:47:41.397
the the nature of the computing. I think
you need a computing that is now geared

00:47:41.480 --> 00:47:45.077
towards real-time context
and from silicon to

00:47:45.160 --> 00:47:48.757
cloud. I mean, this is sort of one
of the reference designs we have which I

00:47:48.840 --> 00:47:53.437
love. There's stuff happening at the edge
here on this device. There's things

00:47:53.520 --> 00:47:55.197
on the cloud. You want to talk a a bit

00:47:55.280 --> 00:47:58.157
about the core systems and silicon

00:47:58.240 --> 00:48:01.517
implications for such a new ecosystem.

00:48:01.600 --> 00:48:06.197
>> You need a very power efficient CPU.
The whole silicon is designed for you to

00:48:06.280 --> 00:48:10.357
have a cloud native experience. And then
you have a lot of sensors. A lot of

00:48:10.440 --> 00:48:15.397
sensors for context. It's a much more
personalized and bespoke experience than

00:48:15.480 --> 00:48:17.197
an app in itself.

00:48:17.280 --> 00:48:19.317
And I think that's changing the nature

00:48:19.400 --> 00:48:22.517
of devices. And even the definition of a

00:48:22.600 --> 00:48:24.957
wearable platform is changing in itself.

00:48:25.040 --> 00:48:28.797
And you started to see those incredible
new form factors. That computing

00:48:28.880 --> 00:48:32.837
platform, one of the things that you
and I have talked a lot about is

00:48:32.920 --> 00:48:35.077
how do you build it such that there is

00:48:35.160 --> 00:48:40.517
an open ecosystem, right? Because it's
not about one agent. It's about any

00:48:40.600 --> 00:48:44.037
agent. >> This is what makes it very, very

00:48:44.120 --> 00:48:45.957
interesting and exciting. Because the

00:48:46.040 --> 00:48:48.717
smartphone today is at the center of

00:48:48.800 --> 00:48:51.357
your digital life.

00:48:51.440 --> 00:48:54.597
The job of those devices today is to be

00:48:54.680 --> 00:48:58.517
extending the functionality of the
smartphone. And because of that, you

00:48:58.600 --> 00:49:02.557
actually saw that many of those
platforms became vertical. It was a

00:49:02.640 --> 00:49:06.757
natural thing to have a vertical
platform from the same company because

00:49:06.840 --> 00:49:10.637
the phone was at the center.
That changes dramatically >> Right.

00:49:10.720 --> 00:49:13.037
>> when you think about agents.

00:49:13.120 --> 00:49:16.557
And agents becomes the center of your

00:49:16.640 --> 00:49:18.437
digital experience. And I think the

00:49:18.520 --> 00:49:20.997
industry is going to be looking for an

00:49:21.080 --> 00:49:23.877
open horizontal platform that actually

00:49:23.960 --> 00:49:27.557
enable the agents to be interacting with

00:49:27.640 --> 00:49:31.677
the best possible device for different
applications. And I think that's what's

00:49:31.760 --> 00:49:34.037
exciting. >> And we want
to make it possible for

00:49:34.120 --> 00:49:35.837
anybody who's thinking of an agentic

00:49:35.920 --> 00:49:38.917
system to not just be bound to current

00:49:39.000 --> 00:49:41.677
devices, but to imagine that there can

00:49:41.760 --> 00:49:46.357
be many, many devices that carry that
intelligence in different contexts. And

00:49:46.440 --> 00:49:50.277
so that's the open ecosystem we want to
build together. And again, it's so great

00:49:50.360 --> 00:49:52.757
to be partnered with you
to get this started.

00:49:52.840 --> 00:49:54.477
>> That's awesome.
We're very proud of this

00:49:54.560 --> 00:49:58.200
partnership and it's just the beginning.

00:49:58.620 --> 00:50:00.640
>> [applause]

00:50:02.400 --> 00:50:05.917
>> Thank you very much Stevie
and Cristiano. We are very excited about

00:50:06.000 --> 00:50:09.557
Project Solara.
It's sort of a new platform

00:50:09.640 --> 00:50:11.677
but perhaps more importantly it's a set

00:50:11.760 --> 00:50:15.437
of new platform rules that don't you

00:50:15.520 --> 00:50:18.237
know, in some sense hem in what you can

00:50:18.320 --> 00:50:19.957
imagine

00:50:20.040 --> 00:50:22.517
the type of form factors, where your

00:50:22.600 --> 00:50:25.397
agents live. I think always whenever

00:50:25.480 --> 00:50:27.717
these new platforms have come,

00:50:27.800 --> 00:50:30.837
you get to rewrite even the rules of how

00:50:30.920 --> 00:50:33.637
new platforms operate and that's what

00:50:33.720 --> 00:50:35.117
we're trying to get done with Project

00:50:35.200 --> 00:50:37.997
Solara so that you as developers

00:50:38.080 --> 00:50:41.757
and and enterprises have the flexibility

00:50:41.840 --> 00:50:44.237
to imagine the form factors that you

00:50:44.320 --> 00:50:47.397
want and have your agents be ubiquitous.

00:50:47.480 --> 00:50:51.597
Now let's go up the stack
to the next layer.

00:50:51.680 --> 00:50:55.437
>> [laughter] >> We are
building a new intelligence layer

00:50:55.520 --> 00:50:58.357
bringing together the models, contexts

00:50:58.440 --> 00:51:03.517
as well as the tools.
It of course starts with model choice.

00:51:03.600 --> 00:51:07.757
Every customer, every
developer is going to choose the

00:51:07.840 --> 00:51:10.157
right model for the right task and their

00:51:10.240 --> 00:51:13.757
eval mix, latency budget, even cogs

00:51:13.840 --> 00:51:15.277
budgets

00:51:15.360 --> 00:51:18.997
and Foundry today has over 11,000

00:51:19.080 --> 00:51:22.677
models. It's the largest model catalog

00:51:22.760 --> 00:51:26.637
out there from Open AI
to Anthropic and of course

00:51:26.720 --> 00:51:29.117
even MAI models and just

00:51:29.200 --> 00:51:31.437
just last week we brought the Open AI

00:51:31.520 --> 00:51:34.437
real-time voice models along with Claude

00:51:34.520 --> 00:51:37.877
Opus 4.8 to Foundry. So we're continuing

00:51:37.960 --> 00:51:40.317
to bring all these frontier models.

00:51:40.400 --> 00:51:43.637
The the consideration
though about the models

00:51:43.720 --> 00:51:46.277
is becoming now increasingly key. Key in

00:51:46.360 --> 00:51:47.997
fact from the last developer conference

00:51:48.080 --> 00:51:50.677
to now, uh when you're building any

00:51:50.760 --> 00:51:53.957
agentic system, having this context,

00:51:54.040 --> 00:51:57.157
uh really shaped right, is becoming

00:51:57.240 --> 00:52:01.117
super important. And in fact, it starts
right at the data tier, right? The data

00:52:01.200 --> 00:52:03.917
estate today has been built for

00:52:04.000 --> 00:52:08.757
applications that supported these
user-facing applications, right? Now you

00:52:08.840 --> 00:52:10.637
have to change and build them for

00:52:10.720 --> 00:52:13.277
agents, which again have very different

00:52:13.360 --> 00:52:16.157
call patterns even to the data tier. Um

00:52:16.240 --> 00:52:18.357
agents are continuously

00:52:18.440 --> 00:52:21.917
storing, retrieving, reasoning,

00:52:22.000 --> 00:52:26.037
uh acting, and learning, right? That's
sort of what's happening in a continuous

00:52:26.120 --> 00:52:28.317
loop. Uh and you see that today. Uh

00:52:28.400 --> 00:52:30.917
agents are using things like Cosmos DB

00:52:31.000 --> 00:52:33.317
for their memory. In fact, ChatGPT does.

00:52:33.400 --> 00:52:36.237
Azure Search, um you you know, is used

00:52:36.320 --> 00:52:39.397
for retrieval of indices and embeddings.

00:52:39.480 --> 00:52:43.917
Uh Fabric IQ for the semantic models
and the ontologies. It's part of the

00:52:44.000 --> 00:52:47.277
business logic tier
effectively for agents.

00:52:47.360 --> 00:52:50.997
Um and also, uh you have Fabric
Real-Time Intelligence for all the

00:52:51.080 --> 00:52:54.117
observability traces. Uh which now

00:52:54.200 --> 00:52:57.157
brings me uh to a very exciting new

00:52:57.240 --> 00:53:00.757
service, Horizon DB, which is our fully

00:53:00.840 --> 00:53:03.397
managed PostgreSQL service on Azure.

00:53:03.480 --> 00:53:06.880
Really thrilled to have this.

00:53:07.920 --> 00:53:10.397
>> [applause] >> I mean,
one of the things we wanted to

00:53:10.480 --> 00:53:12.917
make sure is we built ground up a

00:53:13.000 --> 00:53:15.157
PostgreSQL managed service which was for

00:53:15.240 --> 00:53:18.677
high availability, scale out. Uh it's

00:53:18.760 --> 00:53:22.117
zone redundant with automated failover.

00:53:22.200 --> 00:53:25.997
Uh 128 terabytes of storage per cluster.

00:53:26.080 --> 00:53:28.157
Uh 15 read replicas. I mean, the

00:53:28.240 --> 00:53:31.317
read-heavy workloads, uh you can scale

00:53:31.400 --> 00:53:35.637
uh with this managed service.
And in our internal testing, uh we're

00:53:35.720 --> 00:53:37.557
seeing something like, you know, the

00:53:37.640 --> 00:53:40.917
Horizon is delivering 3x throughput uh

00:53:41.000 --> 00:53:43.357
compared to any self-managed PostgreSQL

00:53:43.440 --> 00:53:47.957
setup. So, it's super critical when you
think about the uh the scale you need.

00:53:48.040 --> 00:53:51.317
Uh the other uh data
workload uh we're also

00:53:51.400 --> 00:53:53.197
changing pretty dramatically is the data

00:53:53.280 --> 00:53:56.717
warehouse. Uh in a world where agents

00:53:56.800 --> 00:53:59.757
are constantly querying data, the data

00:53:59.840 --> 00:54:03.757
warehouse effectively becomes a pretty
mission-critical, right? I mean, it's

00:54:03.840 --> 00:54:05.837
one thing to have it be mission-critical

00:54:05.920 --> 00:54:08.877
for users, uh but when agents need uh

00:54:08.960 --> 00:54:11.477
the analysis done on the fly, uh you

00:54:11.560 --> 00:54:13.917
know, bringing GPU acceleration to

00:54:14.000 --> 00:54:17.237
Fabric uh is super key, and we're seeing

00:54:17.320 --> 00:54:21.477
7x uh performance gain.
So, it's really thrilling

00:54:21.560 --> 00:54:23.817
uh to see that AI acceleration.

00:54:23.900 --> 00:54:25.920
>> [applause]

00:54:26.840 --> 00:54:29.397
>> So, if you sort of have the data tier,

00:54:29.480 --> 00:54:31.797
the the layer above this

00:54:31.880 --> 00:54:35.957
is the IQ layer that we're building that

00:54:36.040 --> 00:54:41.117
brings together essentially
the model capabilities to

00:54:41.200 --> 00:54:44.597
along with the data, right? So,
you kind of mix the data and the model

00:54:44.680 --> 00:54:47.237
capabilities so that you can deliver

00:54:47.320 --> 00:54:51.797
that right context to unlock
intelligence, right? That's where In

00:54:51.880 --> 00:54:56.837
fact, if you want if when people talk
about token efficiency, this is perhaps

00:54:56.920 --> 00:54:58.557
the most important consideration, right?

00:54:58.640 --> 00:55:01.957
If you structure the context right uh

00:55:02.040 --> 00:55:04.677
and feed the models, then you will by

00:55:04.760 --> 00:55:06.317
definition going to be so much more

00:55:06.400 --> 00:55:10.077
token efficient. Uh the first domain uh

00:55:10.160 --> 00:55:12.997
is the web. Uh web grounding is so

00:55:13.080 --> 00:55:16.557
important. Um you need that fresh, uh

00:55:16.640 --> 00:55:21.037
high-quality, and fast um web data, uh

00:55:21.120 --> 00:55:23.157
and that's why we're really, really

00:55:23.240 --> 00:55:28.840
excited uh to announce today WebIQ.

00:55:29.400 --> 00:55:33.957
>> [applause] >> And WebIQ
is built on our global

00:55:34.040 --> 00:55:36.197
infrastructure that's already serving

00:55:36.280 --> 00:55:39.637
over a billion users, but fundamentally

00:55:39.720 --> 00:55:41.277
re-architected

00:55:41.360 --> 00:55:43.477
for the LLM and the agentic workflow.

00:55:43.560 --> 00:55:47.437
It's model agnostic, it's MCP native,

00:55:47.520 --> 00:55:50.757
plugs right into any agent runtime. It

00:55:50.840 --> 00:55:54.717
has web, it has news, images, video. So

00:55:54.800 --> 00:55:58.477
agents can ground responses in fresh,

00:55:58.560 --> 00:56:00.837
verifiable content.

00:56:00.920 --> 00:56:04.157
And Web IQ leads across all of the three

00:56:04.240 --> 00:56:06.517
key criteria, right? It's best in class

00:56:06.600 --> 00:56:09.397
in quality, it's best in class in speed,

00:56:09.480 --> 00:56:13.597
as well as in cost. So we're very,
very thrilled about Web IQ being in the

00:56:13.680 --> 00:56:17.477
developers' hands as you build out
agentic systems. Of course, we're not

00:56:17.560 --> 00:56:19.997
stopping there. Beyond the web, every

00:56:20.080 --> 00:56:21.957
developer wants to be able to ground

00:56:22.040 --> 00:56:25.077
their agents on what's the most valuable

00:56:25.160 --> 00:56:28.117
data across the enterprise. And so we

00:56:28.200 --> 00:56:31.877
are bringing together Foundry, Fabric,

00:56:31.960 --> 00:56:35.637
and Microsoft 365 as this unified IQ

00:56:35.720 --> 00:56:37.997
layer, right? It's a continuously

00:56:38.080 --> 00:56:40.837
updated understanding

00:56:40.920 --> 00:56:43.957
of your organization to show you all of

00:56:44.040 --> 00:56:47.237
this rich IQ in action, building

00:56:47.320 --> 00:56:49.717
real-life agents. Here is Elijah.

00:56:49.800 --> 00:56:54.610
Elijah, take it away. >> [applause]

00:56:55.520 --> 00:56:58.637
>> Agents are only as good as the context

00:56:58.720 --> 00:57:02.077
we give them. Microsoft IQ unifies

00:57:02.160 --> 00:57:06.237
enterprise intelligence
for every organization.

00:57:06.320 --> 00:57:08.357
I'm here at a power utilities control

00:57:08.440 --> 00:57:10.757
center and we'll start by running a

00:57:10.840 --> 00:57:14.197
long-running agent.

00:57:14.280 --> 00:57:16.357
This agent is going to help us assess

00:57:16.440 --> 00:57:21.757
the current grid operations incident and
produce a brief for us so we can respond

00:57:21.840 --> 00:57:24.437
accordingly. Now, I'm going
to go ahead and kick that

00:57:24.520 --> 00:57:28.557
off, and while that runs, let me show
you how we got here with context from

00:57:28.640 --> 00:57:30.757
Microsoft IQ.

00:57:30.840 --> 00:57:33.637
We built our agent in Microsoft Foundry.

00:57:33.720 --> 00:57:36.037
It's connected to various tools, and

00:57:36.120 --> 00:57:38.317
it's also wired to a Foundry IQ

00:57:38.400 --> 00:57:41.277
knowledge base, a single grounded source

00:57:41.360 --> 00:57:46.517
that packages our documents, operational
data, and people into context the agent

00:57:46.600 --> 00:57:50.357
can reason over.
After building the agent in Foundry, we

00:57:50.440 --> 00:57:53.317
published it to Microsoft 365 for the

00:57:53.400 --> 00:57:57.557
whole team. Let's go see it in action.

00:57:57.640 --> 00:58:01.557
Now, here in M365, I will start with
a question about current events that an

00:58:01.640 --> 00:58:03.477
ungrounded LLM wouldn't be able to

00:58:03.560 --> 00:58:05.997
answer. I'm going to ask about current

00:58:06.080 --> 00:58:08.917
electricity prices in SF.

00:58:09.000 --> 00:58:11.157
For this, our agent pulls in the first

00:58:11.240 --> 00:58:14.397
IQ in our toolbox, Web IQ, search built

00:58:14.480 --> 00:58:16.077
for the AI era.

00:58:16.160 --> 00:58:18.797
Web IQ delivers industry-leading

00:58:18.880 --> 00:58:22.317
quality, velocity, and efficiency. Web

00:58:22.400 --> 00:58:25.317
IQ constantly indexes fresh, official

00:58:25.400 --> 00:58:29.677
sources from across the web,
and additionally, Web IQ does a great job

00:58:29.760 --> 00:58:32.157
with semantic documents. And here we can

00:58:32.240 --> 00:58:37.037
see Web IQ gave us our answer
grounded in reality.

00:58:37.120 --> 00:58:41.637
Now, not only can we deploy our agent
here in M365, but we can also embed it

00:58:41.720 --> 00:58:45.517
in our custom apps.
Let's head back to the control center.

00:58:45.600 --> 00:58:49.917
Let's see how we handled a previous
incident. So, after using Web IQ to

00:58:50.000 --> 00:58:54.797
gather external info, we asked
for details about our potentially at-risk

00:58:54.880 --> 00:58:59.037
substations. The real
power of Microsoft IQ comes

00:58:59.120 --> 00:59:01.317
when we combine that external knowledge

00:59:01.400 --> 00:59:03.237
with our own internal enterprise

00:59:03.320 --> 00:59:06.157
context. For this information, our agent

00:59:06.240 --> 00:59:08.677
pulls in the next layer of Microsoft IQ,

00:59:08.760 --> 00:59:11.157
Fabric IQ.

00:59:11.240 --> 00:59:13.757
Here is Brightlines grid represented as

00:59:13.840 --> 00:59:17.077
a fabric ontology, an operational model

00:59:17.160 --> 00:59:19.677
of the live grid. And critically, we

00:59:19.760 --> 00:59:21.837
didn't build this from scratch.

00:59:21.920 --> 00:59:24.197
Fabric takes the Power BI semantic

00:59:24.280 --> 00:59:26.357
models used by millions of customers

00:59:26.440 --> 00:59:31.237
today and let's teams extend them
into rich ontologies that help run the

00:59:31.320 --> 00:59:35.157
business. This model Yes.
This model is coupled

00:59:35.240 --> 00:59:39.877
with live telemetry, so it reflects
the real operational state of the grid

00:59:39.960 --> 00:59:41.877
minute by minute.

00:59:41.960 --> 00:59:47.400
This is what Microsoft IQ means
by enterprise intelligence.

00:59:48.130 --> 00:59:52.237
>> [applause] >> Not data
scattered across disparate

00:59:52.320 --> 00:59:56.597
systems, but a single living model
of the business that an agent can reason

00:59:56.680 --> 01:00:00.157
over. Now, turning back to our agent,

01:00:00.240 --> 01:00:02.797
we can see that it gave us a table of

01:00:02.880 --> 01:00:07.437
our most exposed substations. That's
the agent querying the ontology you just

01:00:07.520 --> 01:00:10.197
saw. Now, the agent has access to both the

01:00:10.280 --> 01:00:12.717
outside world and our grid. The last

01:00:12.800 --> 01:00:18.037
piece is the one that turns a situation
into a response, our policies and our

01:00:18.120 --> 01:00:21.317
people. By asking, "What are the steps to

01:00:21.400 --> 01:00:26.197
respond to a substation trip?" we
activate the final layer of Microsoft IQ

01:00:26.280 --> 01:00:30.557
Work IQ. This is Brightline's
response procedure

01:00:30.640 --> 01:00:35.037
in SharePoint. It's the playbook the team
actually reaches for when something

01:00:35.120 --> 01:00:38.117
goes wrong.
And the important thing is the agent

01:00:38.200 --> 01:00:40.357
isn't working from a stale upload or

01:00:40.440 --> 01:00:46.037
copied snapshot. It's answering from
the same source the team maintains day to

01:00:46.120 --> 01:00:48.917
day. When the procedure
changes, the answer

01:00:49.000 --> 01:00:51.437
changes with it. No re-uploads, no

01:00:51.520 --> 01:00:53.957
prompt rebuilds, and no stale versions.

01:00:54.040 --> 01:00:57.517
And critically, this is your knowledge.

01:00:57.600 --> 01:01:02.677
The assets you create stay with you,
no matter what model or agent is reasoning

01:01:02.760 --> 01:01:06.157
over them. Now, if we
return to our agent, we can

01:01:06.240 --> 01:01:08.197
see the proper way to respond to this

01:01:08.280 --> 01:01:11.037
issue. Not a generic recommendation, but

01:01:11.120 --> 01:01:14.357
our playbook applied to this incident.

01:01:14.440 --> 01:01:19.037
Now, we just saw one situation,
three questions, and one connected answer.

01:01:19.120 --> 01:01:22.277
Now, let's go check back
in on our long-running agent.

01:01:22.360 --> 01:01:24.077
And Oh.

01:01:24.160 --> 01:01:27.357
Let's check the backup really quick.

01:01:27.440 --> 01:01:31.997
Moment of truth. And

01:01:32.080 --> 01:01:38.080
boom. OUR TASK FINISHED.

01:01:38.830 --> 01:01:43.517
>> [applause] >> HERE WE CAN
SEE every step the agent took.

01:01:43.600 --> 01:01:45.717
First, beginning with Web IQ, connecting

01:01:45.800 --> 01:01:48.917
it to the outside world. Second, Fabric

01:01:49.000 --> 01:01:53.277
IQ through Foundry, anchoring it
in the real estate of our operations. And

01:01:53.360 --> 01:01:57.677
third, Work IQ, grounding it
in our people and procedures.

01:01:57.760 --> 01:02:00.997
I triggered this manually, but with
Foundry routines, this can run on a

01:02:01.080 --> 01:02:03.277
schedule, turning a one-off response

01:02:03.360 --> 01:02:06.557
into continuous proactive execution. And

01:02:06.640 --> 01:02:11.117
if we look closely, we can see
it even use the power of Work

01:02:11.200 --> 01:02:14.957
IQ to alert me directly about
the incident. And if we go ahead and check

01:02:15.040 --> 01:02:17.957
Teams, it sent me an incident brief

01:02:18.040 --> 01:02:22.360
notifying me of the situation.

01:02:23.370 --> 01:02:27.997
>> [applause] >> That's
the power of Microsoft IQ. When a

01:02:28.080 --> 01:02:32.357
crisis hits, the team doesn't chase
answers across a dozen systems. They ask

01:02:32.440 --> 01:02:36.437
a question and get a response grounded
in the world, their operations, and

01:02:36.520 --> 01:02:42.160
their people, all in one place
they can trust. Back to you, Satya.

01:02:43.020 --> 01:02:45.957
>> [applause] >> All right.

01:02:46.040 --> 01:02:47.837
Thank you, Elijah.

01:02:47.920 --> 01:02:50.357
Uh and now, you know,

01:02:50.440 --> 01:02:53.037
let's sort of move up from this context

01:02:53.120 --> 01:02:56.557
model layer uh to deploying these agents

01:02:56.640 --> 01:03:00.557
and thinking about the runtime.
Um and you know, when you're building a

01:03:00.640 --> 01:03:04.557
first-class um agentic system,
you need a first-class

01:03:04.640 --> 01:03:07.237
agent runtime and a platform uh that we

01:03:07.320 --> 01:03:09.357
are going to ship both with Windows as

01:03:09.440 --> 01:03:12.317
well as part of Foundry in Azure. Uh we

01:03:12.400 --> 01:03:16.037
want Windows to be a fantastic place to

01:03:16.120 --> 01:03:18.517
run and scale agents.

01:03:18.600 --> 01:03:21.477
Um and agents effectively are a new

01:03:21.560 --> 01:03:23.597
execution environment, right? It's a new

01:03:23.680 --> 01:03:27.077
paradigm even. They reason continuously.

01:03:27.160 --> 01:03:30.117
They generate and run code dynamically.

01:03:30.200 --> 01:03:33.757
Uh they take actions uh across files and

01:03:33.840 --> 01:03:37.077
devices as well as across the network.

01:03:37.160 --> 01:03:40.077
Um obviously there's a lot of power uh

01:03:40.160 --> 01:03:44.117
in it, right? The fact that it can
generate code and act on it on a on a

01:03:44.200 --> 01:03:46.557
long-running agent that's autonomous.

01:03:46.640 --> 01:03:50.917
But obviously it creates new risk.
Uh and that's why today uh we're

01:03:51.000 --> 01:03:52.957
introducing Microsoft Execution

01:03:53.040 --> 01:03:56.797
Containers or MXC. Uh MXC is a new

01:03:56.880 --> 01:04:00.397
policy layer that lets Windows apply

01:04:00.480 --> 01:04:04.637
isolation and containment uh using AI

01:04:04.720 --> 01:04:09.197
native or rather oh sorry, OS native
primitives, right? You need to bake this

01:04:09.280 --> 01:04:11.597
into the operating system. Uh so that

01:04:11.680 --> 01:04:14.597
the containment is enforced by policy.

01:04:14.680 --> 01:04:17.477
Um you can have process-level isolation

01:04:17.560 --> 01:04:19.717
for lightweight agent actions. You can

01:04:19.800 --> 01:04:22.397
have session-level isolation for user

01:04:22.480 --> 01:04:25.237
separation. Uh so Windows and Linux

01:04:25.320 --> 01:04:27.237
virtual machines in fact are also great

01:04:27.320 --> 01:04:30.437
including WSL for much uh stronger

01:04:30.520 --> 01:04:33.157
boundaries. Uh and in fact if you want

01:04:33.240 --> 01:04:38.237
full isolation and containment, Windows
365 for agents for maximum isolation in

01:04:38.320 --> 01:04:39.717
a separate managed environment

01:04:39.800 --> 01:04:42.677
effectively. Uh you can pick the right

01:04:42.760 --> 01:04:45.397
containment option for the workload and

01:04:45.480 --> 01:04:48.277
Windows will enforce it uh via MXC. And

01:04:48.360 --> 01:04:50.957
so I think this becomes pretty critical

01:04:51.040 --> 01:04:52.717
uh as you think about deploying agents

01:04:52.800 --> 01:04:56.437
at scale on your Windows desktop. Uh we

01:04:56.520 --> 01:05:01.477
want to ensure containment is enforced,
of course, regardless of who builds the

01:05:01.560 --> 01:05:05.117
agent, right? So this is why you want to
bake it into the operating system. Uh to

01:05:05.200 --> 01:05:09.237
that end we're working with many
partners to ensure that containment we

01:05:09.320 --> 01:05:11.717
are building supports real developer

01:05:11.800 --> 01:05:13.917
workloads out there and and also

01:05:14.000 --> 01:05:16.677
addresses the needs. In fact, NVIDIA is

01:05:16.760 --> 01:05:18.917
bringing Open Shell to Windows to

01:05:19.000 --> 01:05:22.837
securely execute autonomous AI agents.

01:05:22.920 --> 01:05:27.277
And today we're really
thrilled to announce that

01:05:27.360 --> 01:05:32.720
Open Claw runs on Windows leveraging MXC.

01:05:33.320 --> 01:05:35.340
>> [applause]

01:05:35.520 --> 01:05:37.917
>> Yeah, we are very deeply engaged with

01:05:38.000 --> 01:05:40.797
the team to make Open Claw run super

01:05:40.880 --> 01:05:43.197
well on Windows. And so let me hand it

01:05:43.280 --> 01:05:48.717
over to my colleagues Scott and Samantha
to show you Open Claw on Windows. Scott,

01:05:48.800 --> 01:05:53.030
Samantha. >> [applause] [cheering]

01:05:55.400 --> 01:05:59.877
>> Hey friends. You know, Open Claw came
out in November of last year and it took

01:05:59.960 --> 01:06:03.717
the world by storm. And for the last
several months, I've been using my Open

01:06:03.800 --> 01:06:05.797
Claw to stay on top of my health.

01:06:05.880 --> 01:06:08.037
My Claw can help me manage my blood

01:06:08.120 --> 01:06:10.437
sugar and it gives me even proactive

01:06:10.520 --> 01:06:15.237
notifications via heartbeat. I've got
mine triaging my personal email, it's

01:06:15.320 --> 01:06:18.717
doing my GitHub issues and tracking
packages and it even buys me movie

01:06:18.800 --> 01:06:21.637
tickets. So Samantha,
what are you using your Claw for?

01:06:21.720 --> 01:06:26.597
>> That's cool, Scott. I turned my Open
Claw agent into my triathlon coach. I'm

01:06:26.680 --> 01:06:30.357
nowhere near ready for my race
in September, but coach Claw developed a

01:06:30.440 --> 01:06:35.117
work back plan for me and is using
my Strava data to notify me on how I'm

01:06:35.200 --> 01:06:38.517
progressing and to keep me
accountable when I'm slacking.

01:06:38.600 --> 01:06:41.877
>> Now we've both found our Claws
to be super useful and that's why we're

01:06:41.960 --> 01:06:46.197
working closely with Open Claw to make
them successful and even more successful

01:06:46.280 --> 01:06:50.237
on Windows. We've been collaborating
in the open on GitHub to bring you all an

01:06:50.320 --> 01:06:54.437
Open Claw Windows companion app. It's
going to help you set up your own Claws

01:06:54.520 --> 01:06:58.837
or connect to existing ones whether
they're hosted in Windows or in WSL. And

01:06:58.920 --> 01:07:00.357
the Windows companion, we're going to

01:07:00.440 --> 01:07:03.357
sandbox the Open Claw tool calls to keep

01:07:03.440 --> 01:07:07.637
you and your system safe. >> Yeah,
you'll see the Open Claw Windows

01:07:07.720 --> 01:07:11.157
Companion app running right now in the
background. Go ahead and right-click on

01:07:11.240 --> 01:07:14.640
it, Scott. >> All right.

01:07:15.320 --> 01:07:18.637
>> That looks awesome. You'll notice
immediately it looks like a native

01:07:18.720 --> 01:07:23.237
Windows app because it is. It's written
in WinUI 3. It's got all kinds of

01:07:23.320 --> 01:07:26.917
information about my gateway, other
machines that are participating in my

01:07:27.000 --> 01:07:30.077
claw, my sessions, and my usage. I've

01:07:30.160 --> 01:07:35.117
also got quick access to things like
chat, canvas, the main dashboard, and

01:07:35.200 --> 01:07:39.237
more. Let's jump into companion settings.

01:07:39.320 --> 01:07:43.197
Within the app, we've got full chat
support with tool calling. And you'll

01:07:43.280 --> 01:07:45.277
notice down here in the corner,

01:07:45.360 --> 01:07:50.317
we've got lots of permissions options
along with our sandbox configuration.

01:07:50.400 --> 01:07:54.317
>> Now, this sandbox is really interesting
because this is using MXC, the Microsoft

01:07:54.400 --> 01:07:58.237
Execution Containers, and for this we're
going to be using process isolation.

01:07:58.320 --> 01:08:01.597
Now, newer versions of Windows are going
to have even more containment options,

01:08:01.680 --> 01:08:04.237
so you're going to want to keep an eye
out for more news with MXC in the

01:08:04.320 --> 01:08:06.317
future. Now, I can see that I've got

01:08:06.400 --> 01:08:11.477
one-click security option settings,
but Samantha, talk to me about custom

01:08:11.560 --> 01:08:14.837
folders. >> Yeah, you've
got full support about what

01:08:14.920 --> 01:08:19.157
files and folders you want Open Claw
to have access to, and really granular

01:08:19.240 --> 01:08:21.357
security features like clipboard access

01:08:21.440 --> 01:08:23.837
or talking to the internet itself.

01:08:23.920 --> 01:08:27.117
Now, I've given it read-only desktop

01:08:27.200 --> 01:08:29.397
read access to your desktop folder.

01:08:29.480 --> 01:08:33.837
Open Claw already has a rich safety layer,
and that layer is only augmented

01:08:33.920 --> 01:08:35.837
more by appropriate containment that can

01:08:35.920 --> 01:08:38.477
be managed by me or policies applied by

01:08:38.560 --> 01:08:43.517
IT. Now, for the purposes of this demo,
I'm going to do something really scary

01:08:43.600 --> 01:08:46.877
and ask Open Claw to delete
all the files on your desktop.

01:08:46.960 --> 01:08:49.477
>> That's cool, cuz I
have a nice clean desktop.

01:08:49.560 --> 01:08:53.997
>> Um I think you're keeping a secret.
You hid all your icons from the audience

01:08:54.080 --> 01:08:57.877
while we were on stage. >> That is >>
I think we need to show the audience who

01:08:57.960 --> 01:09:02.837
you really are. >> So disre- so
disrespectful. I know where everything is.

01:09:02.920 --> 01:09:05.997
>> [laughter] >> Just don't
touch my stuff. I know where

01:09:06.080 --> 01:09:09.757
they are. You know, Sam, I need to make
sure that you're clear that a messy

01:09:09.840 --> 01:09:14.077
desktop is an organized mind. I'm pretty
sure that that's the That's the quote,

01:09:14.160 --> 01:09:17.397
right? So, what we've done is we've
asked Open Claw to delete those files

01:09:17.480 --> 01:09:20.437
from the Windows node, and the only
thing that is going to keep from

01:09:20.520 --> 01:09:23.877
happening is MXC because we've turned
off all of the many layers that Open

01:09:23.960 --> 01:09:29.077
Claw offers, but our IT, in this case
Samantha, has set it to read-only. So,

01:09:29.160 --> 01:09:32.077
it's trying to go and delete all
of those files. We can actually see the

01:09:32.160 --> 01:09:35.317
different attempts where it's going and
deleting and then checking the directory

01:09:35.400 --> 01:09:38.877
and then deleting again because it's
very persistent. It wants these files

01:09:38.960 --> 01:09:42.137
gone, and I want I want them to stay.

01:09:42.220 --> 01:09:44.517
>> [sighs] >> Nope. Nope.

01:09:44.600 --> 01:09:47.517
The read-only sandbox is there. 94 JPEGs

01:09:47.600 --> 01:09:52.117
are still on the desktop. Absolutely.
My desktop icons are safe from Samantha's

01:09:52.200 --> 01:09:54.997
reign of terror. >> Foiled again.

01:09:55.080 --> 01:09:59.597
>> Oh my goodness. So bad.
>> Today we've seen security sandboxing,

01:09:59.680 --> 01:10:04.077
WinUI 3, and open source brought
together all in this alpha release of a

01:10:04.160 --> 01:10:08.717
Windows companion app. We think this app
is a great opportunity to showcase Open

01:10:08.800 --> 01:10:12.037
Claw on Windows, and it's only going
to get better in the coming months.

01:10:12.120 --> 01:10:14.797
>> That is right. And by the way,
I want to note that we're doing all this

01:10:14.880 --> 01:10:17.037
development work on this calm Windows

01:10:17.120 --> 01:10:19.357
development machine with all the tools

01:10:19.440 --> 01:10:22.237
that I love like WSL,

01:10:22.320 --> 01:10:25.117
uh containers, and containment. And I've

01:10:25.200 --> 01:10:29.277
even got GitHub Copilot with multi-model
support ready to go. Now, this is

01:10:29.360 --> 01:10:34.437
literally how the team and I have been
working on the Open Claw app on GitHub.

01:10:34.520 --> 01:10:36.157
Now, people might wonder how all of this

01:10:36.240 --> 01:10:38.957
came together. Turns out that over the

01:10:39.040 --> 01:10:41.517
holidays, I got a DM from this random

01:10:41.600 --> 01:10:44.517
guy on on the internet, and it took me a

01:10:44.600 --> 01:10:49.197
couple of days to get back to him.
Uh and then when I did, I we just had this

01:10:49.280 --> 01:10:52.797
kind of cool idea that
maybe he should come to

01:10:52.880 --> 01:10:55.597
to to Microsoft Build. Now, I think
there's one more person that we all want

01:10:55.680 --> 01:10:58.917
to thank for bringing the next
generation of agents to world. Everybody

01:10:59.000 --> 01:11:02.707
put your hands together for the claw
father himself, Peter Steinberger.

01:11:02.790 --> 01:11:04.810
>> [applause]

01:11:05.800 --> 01:11:09.357
>> Yep. Samantha.

01:11:09.440 --> 01:11:12.637
She's just showing off my secret DMs.

01:11:12.720 --> 01:11:14.117
Um

01:11:14.200 --> 01:11:17.437
I'm so excited to see Open Claw native

01:11:17.520 --> 01:11:20.077
on Windows. You know, watching a claw

01:11:20.160 --> 01:11:26.287
try to delete all your desktop file
and just fail made me really happy.

01:11:26.370 --> 01:11:30.517
>> [laughter] >> Because 6 months
ago that would have totally worked.

01:11:30.600 --> 01:11:32.997
You know, I built Open Claw to have

01:11:33.080 --> 01:11:35.037
access to everything. You know, my

01:11:35.120 --> 01:11:38.517
files, my machines, my chats, always on

01:11:38.600 --> 01:11:42.277
and fully open source.
That's what makes it so powerful.

01:11:42.360 --> 01:11:45.717
And that's what also makes
companies a bit nervous.

01:11:45.800 --> 01:11:50.277
You know, what I kept hearing
was Peter, I love my claw.

01:11:50.360 --> 01:11:54.437
Can I use this at work?
And that's what we spent the last few

01:11:54.520 --> 01:11:57.277
months on with Microsoft, GitHub,

01:11:57.360 --> 01:12:00.877
OpenAI, Nvidia, just to name a few.

01:12:00.960 --> 01:12:05.957
We added observability. We added auto
mode for permissions. We changed how

01:12:06.040 --> 01:12:09.637
access works. It's not all or nothing
anymore. You can pick which folder

01:12:09.720 --> 01:12:11.837
should be read only, which one should be

01:12:11.920 --> 01:12:14.157
writeable or hidden.

01:12:14.240 --> 01:12:21.080
So, here's the news. You can totally
run Open Claw inside YOUR COMPANY NOW.

01:12:21.390 --> 01:12:23.410
>> [applause]

01:12:23.760 --> 01:12:26.837
>> AND WE EVEN MADE THE HARNESS itself a

01:12:26.920 --> 01:12:29.277
plugin. You can bring your own.

01:12:29.360 --> 01:12:32.717
Copilot, Codex, whatever you already

01:12:32.800 --> 01:12:36.677
trust and your rules come right with it.

01:12:36.760 --> 01:12:38.917
And then you put Open Claw on top of it.

01:12:39.000 --> 01:12:41.597
You get persistent memory, heartbeats,

01:12:41.680 --> 01:12:44.437
and you get a claw right inside Slack or

01:12:44.520 --> 01:12:49.197
Teams. It's been really
exciting to see Open

01:12:49.280 --> 01:12:50.197
Claw

01:12:50.280 --> 01:12:53.917
grow into something much bigger.

01:12:54.000 --> 01:12:57.277
A global movement and a community.

01:12:57.360 --> 01:13:02.357
And I started the Open Claw Foundation,
a real non-profit, so it stays open and

01:13:02.440 --> 01:13:07.397
neutral. Any model, any operating system.

01:13:07.480 --> 01:13:09.317
Because we are entering

01:13:09.400 --> 01:13:12.597
a new era of building these agents. More

01:13:12.680 --> 01:13:15.677
capability for the people who don't code

01:13:15.760 --> 01:13:18.837
and more power for those who do.

01:13:18.920 --> 01:13:23.517
And we get to do this.
We get to build it together in the open.

01:13:23.600 --> 01:13:28.767
So, my task is simple.
Come build with us. Thank you.

01:13:28.850 --> 01:13:31.677
>> [applause] >> Fantastic.

01:13:31.760 --> 01:13:35.957
Thank you very much to Peter. Thank you
to Samantha. Thank you to the Companion

01:13:36.040 --> 01:13:39.317
app team for their hard work. Thank you
to the Open Claw community for giving

01:13:39.400 --> 01:13:41.197
everyone here a crustacean of their own.

01:13:41.280 --> 01:13:43.957
And as we walk out, I want to remind you
that that's the first time you've ever

01:13:44.040 --> 01:13:47.437
seen Open Claw running
on a Surface Laptop Ultra.

01:13:47.520 --> 01:13:49.870
Goodbye. [cheering]

01:13:50.470 --> 01:13:52.490
>> [applause]

01:13:54.080 --> 01:13:56.877
>> All right. Thank you so much, Samantha,

01:13:56.960 --> 01:13:59.597
Scott, and Peter. It's so wonderful to

01:13:59.680 --> 01:14:03.397
see Open Claw come to Windows

01:14:03.480 --> 01:14:07.317
and have all of that capability in terms

01:14:07.400 --> 01:14:11.277
of the security and and that
comfort to be able to have

01:14:11.360 --> 01:14:15.757
these long-running agents and and
unmetered intelligence come together.

01:14:15.840 --> 01:14:17.997
So, now let's move

01:14:18.080 --> 01:14:20.877
from the Windows side on the edge to the

01:14:20.960 --> 01:14:23.517
cloud with Foundry. And we're building

01:14:23.600 --> 01:14:26.837
Foundry into this full application

01:14:26.920 --> 01:14:29.557
platform for the agent era, right? I

01:14:29.640 --> 01:14:33.117
mean, yeah, every era of the platform

01:14:33.200 --> 01:14:36.877
shift, when we moved to the cloud,
we had the cloud native app stack. And now

01:14:36.960 --> 01:14:40.157
we have the agent native
app stack in Foundry.

01:14:40.240 --> 01:14:42.957
We're particularly excited about the

01:14:43.040 --> 01:14:46.637
Foundry hosted agent Uh, as a runtime

01:14:46.720 --> 01:14:51.357
for long running agents, um, it you know,
agents now have uh, so if you're

01:14:51.440 --> 01:14:55.317
building in Foundry um, uh,
hosted agents, you can have all

01:14:55.400 --> 01:14:59.197
the IQ layers. Uh,
you have the tools. You have the

01:14:59.280 --> 01:15:03.397
durability uh,
and the memory and the state. Uh,

01:15:03.480 --> 01:15:07.597
you have your own sandbox.
In fact, it's a super fast uh,

01:15:07.680 --> 01:15:10.037
sandbox that you can spin up. Uh, you

01:15:10.120 --> 01:15:12.877
can generate the rubrics. Um,

01:15:12.960 --> 01:15:17.397
you get and the evals. Uh,
you can in fact have all the safety and

01:15:17.480 --> 01:15:20.917
the guardrails around your agentic system.

01:15:21.000 --> 01:15:24.637
Uh, and in fact, even one of the coolest
things in Foundry is it's a continuously

01:15:24.720 --> 01:15:28.357
improving loop even, right? It's got
that self-improvement loop built in,

01:15:28.440 --> 01:15:30.077
right? So, you build an agent that's

01:15:30.160 --> 01:15:35.717
continuously getting better. Uh,
and I'm really also excited today

01:15:35.800 --> 01:15:39.957
uh, to announce a partnership
with Fireworks AI,

01:15:40.040 --> 01:15:42.717
uh, bringing in fact all of their open

01:15:42.800 --> 01:15:45.597
weight models to Foundry. Uh, that means

01:15:45.680 --> 01:15:49.877
giving you as developers more choices uh,
as well as their great influence

01:15:49.960 --> 01:15:52.437
stack to build the next generation of

01:15:52.520 --> 01:15:54.637
these agentic applications with all the

01:15:54.720 --> 01:15:58.197
enterprise rails that Foundry has

01:15:58.280 --> 01:16:01.677
uh, as well as the governance uh,
Foundry has. So, really excited about

01:16:01.760 --> 01:16:06.110
that partnership. >> [applause]

01:16:07.120 --> 01:16:10.197
>> And so,
now that brings us to the tools.

01:16:10.280 --> 01:16:15.477
Uh, GitHub itself is at the heart of all
this. Uh, you know, in fact, Jensen

01:16:15.560 --> 01:16:19.397
spoke to this. The GitHub is not just
about the code repo. It's becoming the

01:16:19.480 --> 01:16:23.437
control plane uh, for all
the agents. Um, and nearly

01:16:23.520 --> 01:16:25.757
everything we measure on GitHub uh,

01:16:25.840 --> 01:16:28.637
whether it's repo creation, PR activity,

01:16:28.720 --> 01:16:31.757
API usage, actions, all of them uh, are

01:16:31.840 --> 01:16:33.797
growing faster uh, because of these

01:16:33.880 --> 01:16:37.077
agentic workflows. Uh, this new scale is

01:16:37.160 --> 01:16:39.197
driven by humans and agents

01:16:39.280 --> 01:16:41.317
collaborating together. Uh, and we're

01:16:41.400 --> 01:16:44.637
exposing our tooling across every form

01:16:44.720 --> 01:16:50.997
factor. In fact, we've seen
tremendous growth in CLI driven, you

01:16:51.080 --> 01:16:55.197
know, the approachability of the CLI form
factor. It's always been great to go to

01:16:55.280 --> 01:17:00.037
a terminal. And now though, when combined
with the power of the models and natural

01:17:00.120 --> 01:17:04.317
language, you know, CLI has become the
thing that everybody goes to. But at the

01:17:04.400 --> 01:17:05.837
end of the day,

01:17:05.920 --> 01:17:09.797
when you have hundreds of CLIs, it

01:17:09.880 --> 01:17:13.557
becomes complicated. It doesn't scale.

01:17:13.640 --> 01:17:15.477
Especially the cognitive load that I

01:17:15.560 --> 01:17:19.117
have when you have 100 CLI sessions open

01:17:19.200 --> 01:17:21.277
is such that you kind of need something

01:17:21.360 --> 01:17:23.877
new. And so that's what has led us to

01:17:23.960 --> 01:17:27.517
build We know we need like this tool

01:17:27.600 --> 01:17:30.477
essentially that has the speed and the

01:17:30.560 --> 01:17:33.357
flexibility of a CLI,

01:17:33.440 --> 01:17:35.797
but has the capability of an IDE and the

01:17:35.880 --> 01:17:38.597
ability to scale to infinite number of

01:17:38.680 --> 01:17:40.877
agent sessions. And so today we're

01:17:40.960 --> 01:17:43.997
taking that next big step introducing

01:17:44.080 --> 01:17:48.120
our new GitHub Copilot app.

01:17:49.550 --> 01:17:53.557
>> [applause] >> And we realize
that it's not sufficient,

01:17:53.640 --> 01:17:56.957
right? Because you still have a back
end to deal with, right? Code is easy to

01:17:57.040 --> 01:17:58.637
generate, but what about the back end?

01:17:58.720 --> 01:18:01.117
So you need to contend with identities,

01:18:01.200 --> 01:18:04.277
storage, database schemas. And that's

01:18:04.360 --> 01:18:06.637
why we're really super excited about

01:18:06.720 --> 01:18:08.397
Rayfin. Rayfin

01:18:08.480 --> 01:18:12.477
is an agent first SDK that connects your

01:18:12.560 --> 01:18:15.837
agents to a back end as a service.

01:18:15.920 --> 01:18:18.197
And we're bringing this to everywhere

01:18:18.280 --> 01:18:20.517
you build. That's why I'm again super

01:18:20.600 --> 01:18:23.517
excited about Rayfin and the partnership

01:18:23.600 --> 01:18:30.520
with Replit. You can now build
apps in Replit while the app

01:18:31.120 --> 01:18:33.277
You can build the app in Replit while

01:18:33.360 --> 01:18:38.357
the app and data are deployed into the
enterprise managed fabric tenant thanks

01:18:38.440 --> 01:18:41.997
to, uh, the Ray Fin SDK.
Uh, and so this Ray

01:18:42.080 --> 01:18:46.237
Fin SDK is available now for anyone else,
uh, to be able to use with their

01:18:46.320 --> 01:18:48.397
tools as a back end, uh, and essentially

01:18:48.480 --> 01:18:50.357
have this back end service.

01:18:50.440 --> 01:18:53.357
And now to show you all of the foundry

01:18:53.440 --> 01:18:56.117
and Ray Fin and building agents and the

01:18:56.200 --> 01:19:01.960
long running agents, let me invite up
on stage Cassidy. Cassidy, take it away.

01:19:02.250 --> 01:19:05.517
>> [applause] >> Hello, everyone.

01:19:05.600 --> 01:19:09.317
I'm so excited to be amongst my fellow
devs today. This has been a long day.

01:19:09.400 --> 01:19:11.357
Fix your shoulders. Sit back. Get ready.

01:19:11.440 --> 01:19:13.157
I see a lot of you sitting up. Great.

01:19:13.240 --> 01:19:16.717
Great. I know you're drinking from the
firehose of information today, so I want

01:19:16.800 --> 01:19:20.237
you to go into the next few minutes
thinking, what can I try out on my

01:19:20.320 --> 01:19:23.597
laptop later today?
So as Satya said, we're going to show

01:19:23.680 --> 01:19:28.757
you Ray Fin, but first, I can't wait
to show you the new GitHub Copilot app.

01:19:28.840 --> 01:19:31.837
This app is your home base
for development and operations on your

01:19:31.920 --> 01:19:35.477
computer, and we think you're going
to love it. So let me show you around.

01:19:35.560 --> 01:19:38.277
When you open up the app from the start,
you see this home screen here, where you

01:19:38.360 --> 01:19:42.357
can kick off a new agentic coding session.
But also before I get into the

01:19:42.440 --> 01:19:44.917
serious stuff, you can drag Mona around,

01:19:45.000 --> 01:19:49.277
and there's a game. Look, it's so fun.

01:19:49.360 --> 01:19:53.637
Okay, I'm not very good at it, so let's
just get back to you can kick off a new

01:19:53.720 --> 01:19:55.797
agentic coding session.

01:19:55.880 --> 01:20:00.837
So, I started off one a little bit
earlier here, and it gave me a review of

01:20:00.920 --> 01:20:05.557
a bunch of release blockers.
Which one should I fix? Call it out.

01:20:05.640 --> 01:20:09.197
83? The critical ones? You know what? How

01:20:09.280 --> 01:20:11.997
about we just do all of them?

01:20:12.080 --> 01:20:14.557
Let's go. This app will now kick off a

01:20:14.640 --> 01:20:19.077
separate session for every single issue
here. I don't worry have to worry about

01:20:19.160 --> 01:20:23.397
stashing or coding conflicts or anything
cuz the app takes care of that with Git

01:20:23.480 --> 01:20:26.797
worktrees. Git worktrees
are isolated environments

01:20:26.880 --> 01:20:29.957
for each session that you run, so
your agents can work in parallel without

01:20:30.040 --> 01:20:34.037
stepping on each other.
But you still have to merge them, right?

01:20:34.120 --> 01:20:37.557
So, uh Copilot has
your back there, too. If

01:20:37.640 --> 01:20:40.117
I head over, not to this one, but to

01:20:40.200 --> 01:20:43.637
this issue here, I can run agent merge.

01:20:43.720 --> 01:20:47.837
And when I enable agent merge,
Copilot will continuously babysit this PR

01:20:47.920 --> 01:20:52.397
through CI checks, code review,
and merge conflicts. Okay, let me keep

01:20:52.480 --> 01:20:54.997
showing you around while
those are still running.

01:20:55.080 --> 01:20:57.637
Now, if I head over to my work,

01:20:57.720 --> 01:21:00.357
I can see a focused view of all of my

01:21:00.440 --> 01:21:04.717
activity and just projects loaded in
the app, issues and PRs, everything here.

01:21:04.800 --> 01:21:08.557
And then, under automations, I have
a bunch of reusable sessions and workflows

01:21:08.640 --> 01:21:11.117
that can run locally or on the cloud.

01:21:11.200 --> 01:21:12.797
You see, there's issue poetry there.

01:21:12.880 --> 01:21:15.317
That is real and that is load-bearing.

01:21:15.400 --> 01:21:20.397
Okay. And now, [laughter] under sessions,
like I briefly showed earlier, these are

01:21:20.480 --> 01:21:23.277
sessions. If I want to add
a new repository, I can

01:21:23.360 --> 01:21:27.037
click that button here and it can
pull from a local repo or from a GitHub

01:21:27.120 --> 01:21:29.717
repository. And then, if I were to just

01:21:29.800 --> 01:21:32.677
add one, I can add a session in Pocket

01:21:32.760 --> 01:21:37.037
Cal. This is an open-source repo.
I can start a session anywhere and it just

01:21:37.120 --> 01:21:40.997
loads it. I don't have to clone,
I don't have to pull, it just works.

01:21:41.080 --> 01:21:44.397
Now, I when I look at a session within
this repository, let me look at this

01:21:44.480 --> 01:21:48.597
other one over here, I get an integrated
browser, there's a terminal, I can see

01:21:48.680 --> 01:21:52.677
the chat, it's all loading. I can even
toggle light mode and dark mode in here.

01:21:52.760 --> 01:21:56.917
And there's also this great button, pick
and polish, where if I click that, I can

01:21:57.000 --> 01:21:59.557
pick and polish anything in this app and

01:21:59.640 --> 01:22:04.597
it adds it to the chat and I can say,
"Hey, I want you to add reordering to

01:22:04.680 --> 01:22:08.077
this list." And it'll just
work, all living in there.

01:22:08.160 --> 01:22:11.917
I have access to all the most popular
models via my single GitHub Copilot

01:22:12.000 --> 01:22:16.717
subscription, including those from OpenAI,
Anthropic, and Google. You can

01:22:16.800 --> 01:22:20.517
see all of them in our model picker.
And having model choice is great. Not only

01:22:20.600 --> 01:22:23.797
can you pick the right model for the task,
but for bigger features, Copilot

01:22:23.880 --> 01:22:26.157
can request a rubber duck review. So, in

01:22:26.240 --> 01:22:28.077
this session here, for example, I was

01:22:28.160 --> 01:22:31.877
using GPT-5.5, but if I scroll up, it

01:22:31.960 --> 01:22:35.677
actually requested one
from Claude Opus 4.8.

01:22:35.760 --> 01:22:39.557
All models have blind spots, and the power
of the Copilot multimodal approach

01:22:39.640 --> 01:22:42.197
means that I can catch them earlier.

01:22:42.280 --> 01:22:44.237
And this is all very cool.

01:22:44.320 --> 01:22:49.357
But working with AI in 2026 should
be more than just chat. You just

01:22:49.440 --> 01:22:52.357
saw me scrolling on that.
There's so many words here.

01:22:52.440 --> 01:22:54.717
So, today, I'm very excited to show you

01:22:54.800 --> 01:22:57.517
the concept of a canvas. I'm going to

01:22:57.600 --> 01:22:59.597
open one right there.

01:22:59.680 --> 01:23:01.917
The canvas is how an agent can build a

01:23:02.000 --> 01:23:05.157
custom UI to communicate with you.

01:23:05.240 --> 01:23:08.917
What if your AI could see?

01:23:09.000 --> 01:23:11.477
Everyone say, "Demo gods bless us."

01:23:11.560 --> 01:23:13.877
Okay, let's see if it works.

01:23:13.960 --> 01:23:16.757
Here is a fun canvas where, if I get the

01:23:16.840 --> 01:23:19.677
camera going, okay, the agent shows your

01:23:19.760 --> 01:23:24.237
PRs down here, and I can take toggle it
with a thumbs up or a thumbs down. Let's

01:23:24.320 --> 01:23:29.397
approve it. Hey!
It's so fun! >> [laughter]

01:23:29.480 --> 01:23:32.517
>> And they can go so much deeper if you
want. This is just the beginning of what

01:23:32.600 --> 01:23:35.277
you can do. So, again,
I kicked off a bunch of

01:23:35.360 --> 01:23:38.157
different sessions earlier.

01:23:38.240 --> 01:23:41.557
This is a signal box app. It's 100%

01:23:41.640 --> 01:23:45.477
agent-built. It's containerized
with a database back-end.

01:23:45.560 --> 01:23:50.597
Would you be able to deploy this to your
enterprise with no questions asked? Be

01:23:50.680 --> 01:23:56.237
honest. No. Yes, exactly.
No, but you can with

01:23:56.320 --> 01:23:59.197
Rayfen. And that is very,
very exciting. Let me

01:23:59.280 --> 01:24:02.397
open up a new terminal over here.

01:24:02.480 --> 01:24:07.277
All I have to do is type Rayfen up. And

01:24:07.360 --> 01:24:10.277
then, demo gods bless us. Come on.

01:24:10.360 --> 01:24:12.557
It will

01:24:12.640 --> 01:24:16.117
maybe deploy. Blammo! It's happening.

01:24:16.200 --> 01:24:20.437
Yes! And all hosted on Microsoft Fabric.

01:24:20.520 --> 01:24:23.277
Okay, I know that's a lot.
I know that's a lot.

01:24:23.360 --> 01:24:27.037
With Rayfin, your agents get a complete
enterprise back end, so you can deploy

01:24:27.120 --> 01:24:30.677
with confidence in the way that's best
for you. But, this is the key thing to

01:24:30.760 --> 01:24:33.317
remember. This app

01:24:33.400 --> 01:24:35.437
is not just another session manager.

01:24:35.520 --> 01:24:39.997
Yes, it is. It It manages a lot of
sessions, but session managers just make

01:24:40.080 --> 01:24:44.917
it easy to create work, but GitHub
Copilot helps you to finish it.

01:24:45.000 --> 01:24:47.357
Thank you so much.
Happy Pride. Back to you, Satya.

01:24:47.440 --> 01:24:49.460
>> [applause and cheering]

01:24:52.680 --> 01:24:54.117
[applause]

01:24:54.200 --> 01:24:57.637
>> All right.
Um now we're back to IDs that

01:24:57.720 --> 01:25:00.637
have UI. That's sort of so cool.

01:25:00.720 --> 01:25:02.957
Um you know, it's kind of come so full

01:25:03.040 --> 01:25:07.077
circle. Um now let's talk about how you

01:25:07.160 --> 01:25:10.197
can observe, govern, and secure these

01:25:10.280 --> 01:25:13.917
agents. Uh Agent 365 is the agent uh

01:25:14.000 --> 01:25:17.237
control plane. Uh agent requires

01:25:17.320 --> 01:25:19.717
um their own identities, access

01:25:19.800 --> 01:25:24.357
controls, um even when they're working
on your behalf, right? You just want

01:25:24.440 --> 01:25:26.517
that work on behalf identity uh to be

01:25:26.600 --> 01:25:29.877
enforced. So, we extended in Entra.

01:25:29.960 --> 01:25:33.157
Uh agents need uh real-time defense, so

01:25:33.240 --> 01:25:35.037
we extended Defender.

01:25:35.120 --> 01:25:40.237
Uh agents require this always-on
data protections and compliance, so we

01:25:40.320 --> 01:25:41.997
extended Purview.

01:25:42.080 --> 01:25:44.437
Um and these agents can be hosted

01:25:44.520 --> 01:25:47.677
anywhere. They can be on AWS, GCP, not

01:25:47.760 --> 01:25:49.877
just on Azure, uh or built with any

01:25:49.960 --> 01:25:53.717
framework. Um and today we're announcing

01:25:53.800 --> 01:25:57.077
a number of updates, including the GA of

01:25:57.160 --> 01:26:00.917
Agent 365 SDK, and we're expanding it to

01:26:01.000 --> 01:26:03.317
your local agents uh running on Windows

01:26:03.400 --> 01:26:05.197
and elsewhere, and the clause we just

01:26:05.280 --> 01:26:08.037
saw earlier. And let us take a look at

01:26:08.120 --> 01:26:14.800
how all of Agent 365 composes.
Uh over to you, Amanda.

01:26:16.520 --> 01:26:20.117
>> [applause] >> Everyone
is building agents, but that's

01:26:20.200 --> 01:26:22.677
not the hard part anymore. The hard part

01:26:22.760 --> 01:26:24.717
is integrating them into your business

01:26:24.800 --> 01:26:27.317
and governing them at scale. Today, I'll

01:26:27.400 --> 01:26:30.317
show you how Foundry makes this easy.

01:26:30.400 --> 01:26:34.877
Let's start locally. I've already built
a LangGraph agent, and now I'll show you

01:26:34.960 --> 01:26:39.157
the value Foundry adds. First, tools.

01:26:39.240 --> 01:26:41.517
Agents need tools to get work done, and

01:26:41.600 --> 01:26:43.437
with Foundry toolbox, I just add my

01:26:43.520 --> 01:26:46.277
tools once, and any agent can consume

01:26:46.360 --> 01:26:49.677
them through a single MCP endpoint.

01:26:49.760 --> 01:26:54.357
And because tools live centrally,
that means governance does as well.

01:26:54.440 --> 01:26:56.157
I've applied a guardrail, which blocks

01:26:56.240 --> 01:26:59.037
PII from leaking into tool calls, into

01:26:59.120 --> 01:27:02.237
responses, and all I have to do is apply

01:27:02.320 --> 01:27:06.077
this once, and all
my agents are protected.

01:27:06.160 --> 01:27:09.557
Now, let's make this
agent enterprise-ready.

01:27:09.640 --> 01:27:13.477
To deploy this agent to Foundry,
all I have to do is add this one block of

01:27:13.560 --> 01:27:15.957
code, and then I push my changes, and

01:27:16.040 --> 01:27:18.837
GitHub Actions takes it from there.

01:27:18.920 --> 01:27:23.677
Once deployed, I can actually use
this agent in the same Foundry extension I

01:27:23.760 --> 01:27:25.357
showed you earlier.

01:27:25.440 --> 01:27:29.117
And let's now test it out. I'm asking it

01:27:29.200 --> 01:27:33.277
to track a few open items from
a stand-up I had earlier today,

01:27:33.360 --> 01:27:35.477
and what's actually happening right now

01:27:35.560 --> 01:27:37.917
is Foundry is spinning up a dedicated

01:27:38.000 --> 01:27:41.277
micro VM just for this session. And the

01:27:41.360 --> 01:27:43.997
session even gets its own persistent

01:27:44.080 --> 01:27:46.237
file system. What you're going to see in

01:27:46.320 --> 01:27:48.477
a minute here is if I go to the files

01:27:48.560 --> 01:27:51.957
tab to track the open items, the agent's

01:27:52.040 --> 01:27:54.957
actually writing to a file.

01:27:55.040 --> 01:27:56.597
Pretty cool, right?

01:27:56.680 --> 01:28:00.597
Um plus, Foundry now has server-side

01:28:00.680 --> 01:28:03.117
traces and built-in evals to show me

01:28:03.200 --> 01:28:06.477
exactly what happened on every run.

01:28:06.560 --> 01:28:09.117
But how do I know if my agent's doing a

01:28:09.200 --> 01:28:12.357
good job? That's what Foundry's brand

01:28:12.440 --> 01:28:17.317
rubric evaluators are for.
It's super simple. With just one AZD

01:28:17.400 --> 01:28:20.957
command, Foundry reads my agent and then

01:28:21.040 --> 01:28:23.317
generates the evaluation criteria for

01:28:23.400 --> 01:28:26.437
me. In other words, it creates a rubric

01:28:26.520 --> 01:28:30.797
personalized just to this agent.
Now let's check out this rubric in Foundry

01:28:30.880 --> 01:28:35.997
portal. Look at these
dimensions. Ground uh

01:28:36.080 --> 01:28:40.397
governance outcome correctness,
prescribe source usage. I didn't write

01:28:40.480 --> 01:28:42.837
any of these. Foundry generated them

01:28:42.920 --> 01:28:45.197
from production traces. And with this

01:28:45.280 --> 01:28:48.157
rubric, I can score my agent in run

01:28:48.240 --> 01:28:50.957
evals, but we can do so much more than

01:28:51.040 --> 01:28:53.837
that. That's where Foundry's brand new

01:28:53.920 --> 01:28:56.197
agent optimizer takes over.

01:28:56.280 --> 01:28:58.077
Here's how it works. It tunes four

01:28:58.160 --> 01:29:01.037
things: the model, instructions, tool

01:29:01.120 --> 01:29:03.597
descriptions, and skills. And then it

01:29:03.680 --> 01:29:08.317
generates improved candidates and scores
each one using the rubric I just showed

01:29:08.400 --> 01:29:11.557
you. Here, I can view
the candidates. I can

01:29:11.640 --> 01:29:16.477
see the strategy used. I can see their
scores. And I can actually view exactly

01:29:16.560 --> 01:29:19.677
what changed. This candidate,
for example, improved

01:29:19.760 --> 01:29:23.437
its score by updating
the model and the system prompt.

01:29:23.520 --> 01:29:28.157
Foundry then makes it super easy to deploy
the best candidate as a brand new

01:29:28.240 --> 01:29:30.197
agent version.

01:29:30.280 --> 01:29:33.037
But this is not a one-time thing.

01:29:33.120 --> 01:29:36.157
Every run feeds the next eval, and every

01:29:36.240 --> 01:29:38.357
eval tells the optimizer where to

01:29:38.440 --> 01:29:41.077
improve next. So your agents now get

01:29:41.160 --> 01:29:43.277
better the more they're used.

01:29:43.360 --> 01:29:46.477
But now let's put this agent to work.

01:29:46.560 --> 01:29:49.117
I've published my agent to Teams and

01:29:49.200 --> 01:29:52.357
M365 Copilot. And right now I'm going to

01:29:52.440 --> 01:29:54.717
ask it to just catch me up on what I've

01:29:54.800 --> 01:29:56.797
missed because I've been offline all

01:29:56.880 --> 01:30:00.117
morning, but my agent hasn't.

01:30:00.200 --> 01:30:02.197
While it works on that though, let me

01:30:02.280 --> 01:30:04.877
explain what makes this agent different.

01:30:04.960 --> 01:30:07.437
This is an autopilot agent, which means

01:30:07.520 --> 01:30:09.837
it has its own identity and productivity

01:30:09.920 --> 01:30:13.517
license, so it can work across M365 on

01:30:13.600 --> 01:30:16.237
its own behalf. Earlier, you saw me

01:30:16.320 --> 01:30:19.237
using this agent by myself, but shipping

01:30:19.320 --> 01:30:23.997
a feature in a new release takes a whole
team, and that's why this agent lives in

01:30:24.080 --> 01:30:28.277
our team's group chat. Now, let's
check out how it did. It summarized all

01:30:28.360 --> 01:30:30.837
updates and called out the key features

01:30:30.920 --> 01:30:32.917
I need to be tracking.

01:30:33.000 --> 01:30:35.677
But, to put an agent like this actually

01:30:35.760 --> 01:30:38.197
to work in your enterprise, governance

01:30:38.280 --> 01:30:40.677
needs to come first. And that's why

01:30:40.760 --> 01:30:43.277
every Autopilot agent requires admin

01:30:43.360 --> 01:30:47.797
approval. Here, admins can review all
critical details, and they can even

01:30:47.880 --> 01:30:52.557
choose who has the ability to talk
to the agent. But, it doesn't end there.

01:30:52.640 --> 01:30:54.597
Even after approval,

01:30:54.680 --> 01:30:56.637
uh admins can continue to monitor the

01:30:56.720 --> 01:31:00.957
agent and block it at any time. But,

01:31:01.040 --> 01:31:03.397
governance cannot apply to just one

01:31:03.480 --> 01:31:06.437
agent. Every agent in your organization

01:31:06.520 --> 01:31:08.917
needs to be managed with the same rigor

01:31:09.000 --> 01:31:14.357
as users, apps, and devices. And that's
exactly what Microsoft Agent 365

01:31:14.440 --> 01:31:18.397
provides. So, let's zoom out.
Today, we saw

01:31:18.480 --> 01:31:22.517
Foundry accelerate development,
take your agent from local to

01:31:22.600 --> 01:31:26.757
enterprise-ready,
and put it to work in M365.

01:31:26.840 --> 01:31:30.957
Foundry makes it simple. You build the
agent, we handle the rest. Now, back to

01:31:31.040 --> 01:31:37.520
you, Satya. >> [applause]
>> Thank you, Amanda.

01:31:38.440 --> 01:31:43.157
You know, what you just saw
was how we're building security

01:31:43.240 --> 01:31:46.957
for AI. Uh but, there's
also one other critical

01:31:47.040 --> 01:31:49.237
aspect, especially, you know, the news

01:31:49.320 --> 01:31:52.237
today is all about, you know, how do you

01:31:52.320 --> 01:31:56.757
defend yourself using AI against attacks

01:31:56.840 --> 01:31:59.117
that may, in fact, be using AI, right?

01:31:59.200 --> 01:32:02.117
So, last month, we announced our

01:32:02.200 --> 01:32:03.877
multi-model

01:32:03.960 --> 01:32:07.077
agentic security system M-Dash. That's

01:32:07.160 --> 01:32:09.957
essentially an agent harness for

01:32:10.040 --> 01:32:12.797
security essentially that we built.

01:32:12.880 --> 01:32:15.237
We're bringing together 100 agents

01:32:15.320 --> 01:32:17.837
across the frontier

01:32:17.920 --> 01:32:20.677
and custom models to really find these

01:32:20.760 --> 01:32:25.397
exploitable bugs better than any single
model does. In fact, when we debuted

01:32:25.480 --> 01:32:28.757
this harness,
it was on the top of Cyber Gym

01:32:28.840 --> 01:32:31.237
benchmark. So, I want you to take a look

01:32:31.320 --> 01:32:34.797
at what you can do with M-Dash to really

01:32:34.880 --> 01:32:41.680
defend against AI attacks and defend the
entire digital estate. Over to you, Sarah.

01:32:43.630 --> 01:32:45.650
>> [applause]

01:32:46.360 --> 01:32:51.037
>> Thanks, Satya. Now, we all know that
security scans can take a while, so I'm

01:32:51.120 --> 01:32:53.117
not going to do one live. So, let me

01:32:53.200 --> 01:32:55.397
show you the results of an M-Dash scan I

01:32:55.480 --> 01:32:58.357
already ran on my code base.

01:32:58.440 --> 01:33:00.437
Now, the system runs as a standalone

01:33:00.520 --> 01:33:03.277
CLI, but today I'm using it in my GitHub

01:33:03.360 --> 01:33:06.717
Copilot app on my local dev machine.

01:33:06.800 --> 01:33:09.357
The scan is broken down by vulnerability

01:33:09.440 --> 01:33:14.277
domains and severity. And it's also

01:33:14.360 --> 01:33:16.837
in additional to finding traditional

01:33:16.920 --> 01:33:21.597
issues like coding errors and hardcoded
secrets, it's also identifying AI

01:33:21.680 --> 01:33:23.837
specific vulnerabilities in the code

01:33:23.920 --> 01:33:26.597
base. Now, what happened under the hood

01:33:26.680 --> 01:33:29.077
is that over 100 specialized agents are

01:33:29.160 --> 01:33:32.237
working together to discover, debate,

01:33:32.320 --> 01:33:35.037
and prove exploitable vulnerabilities

01:33:35.120 --> 01:33:37.637
end-to-end. And when the scan finishes,

01:33:37.720 --> 01:33:42.597
it generates both a SARIF log
and an HTML report that I can give to my

01:33:42.680 --> 01:33:46.917
management. Now,
the Defender details command allows

01:33:47.000 --> 01:33:49.317
me to dig into these vulnerabilities,

01:33:49.400 --> 01:33:52.477
and I can see what the vulnerability is,

01:33:52.560 --> 01:33:55.357
where it is in my code, and the severity

01:33:55.440 --> 01:33:58.197
to help me prioritize. And from here, of

01:33:58.280 --> 01:34:01.797
course, we're going to fix it.

01:34:01.880 --> 01:34:04.517
Now, using the Defender fix command,

01:34:04.600 --> 01:34:09.037
the system will remediate suggested
fixes directly in my local dev

01:34:09.120 --> 01:34:13.837
environment. And when that's done,
I can check out the diff. So, I have full

01:34:13.920 --> 01:34:18.597
transparency about what the harness has
done, and I still have a human in the

01:34:18.680 --> 01:34:22.157
loop check. But, of course,

01:34:22.240 --> 01:34:23.957
everything I've shown you so far has run

01:34:24.040 --> 01:34:27.797
locally, but I can also create a PR

01:34:27.880 --> 01:34:30.277
to plug into my existing workflows and

01:34:30.360 --> 01:34:32.797
push up to my repo. And I can take the

01:34:32.880 --> 01:34:35.357
SARIF output from the scan, and I can

01:34:35.440 --> 01:34:38.077
upload it to tools like GitHub Advanced

01:34:38.160 --> 01:34:42.717
Security, and manage everything
alongside my other application security

01:34:42.800 --> 01:34:46.597
findings. Now, I've shown
you m-dash working on my

01:34:46.680 --> 01:34:48.877
code, but let me show you a

01:34:48.960 --> 01:34:51.077
vulnerability our security research

01:34:51.160 --> 01:34:54.197
teams identified using m-dash that you

01:34:54.280 --> 01:34:56.157
can go and look up yourself.

01:34:56.240 --> 01:34:58.357
So, the TLDR, because this is a lot to

01:34:58.440 --> 01:35:02.237
read, of this bug is that Wasmtime reads

01:35:02.320 --> 01:35:04.837
an out-of-date map

01:35:04.920 --> 01:35:07.437
of an object, runs off the end, and then

01:35:07.520 --> 01:35:10.677
it crashes the host. So, this is exactly

01:35:10.760 --> 01:35:12.917
the kind of bug that the harness was

01:35:13.000 --> 01:35:15.597
built for, because the flaw spread

01:35:15.680 --> 01:35:17.717
across three different parts of the code

01:35:17.800 --> 01:35:20.677
base. So, no single file looks wrong on

01:35:20.760 --> 01:35:25.077
its own. They look absolutely fine, and

01:35:25.160 --> 01:35:29.597
we can even see here, which is probably
my favorite part of this bug, a very

01:35:29.680 --> 01:35:34.037
confident statement from the developers
claiming everything is fine.

01:35:34.120 --> 01:35:35.837
Now, this is exactly the sort of

01:35:35.920 --> 01:35:38.437
reassurance that fools normal scanners

01:35:38.520 --> 01:35:41.437
and single AI models, but m-dash wasn't

01:35:41.520 --> 01:35:43.677
fooled. One team of agents spotted the

01:35:43.760 --> 01:35:46.397
suspicious gap, another team argued it

01:35:46.480 --> 01:35:49.437
apart, and a third team built a working

01:35:49.520 --> 01:35:53.717
example that actually triggered the crash.
And it did all of this in an

01:35:53.800 --> 01:35:57.597
open-source code base. And this is
the kind of joined-up reasoning that

01:35:57.680 --> 01:36:00.397
previously required significant manual

01:36:00.480 --> 01:36:02.597
security research effort.

01:36:02.680 --> 01:36:05.197
This is M-Dash helping developers create

01:36:05.280 --> 01:36:08.077
secure code from the start. Coming soon

01:36:08.160 --> 01:36:10.277
to your CLI and the Microsoft Defender

01:36:10.360 --> 01:36:15.317
portal. Back to you, Satya. >> All right.

01:36:15.400 --> 01:36:16.837
Thank you, Sarah.

01:36:16.920 --> 01:36:20.357
So, that was the stack. Before though we

01:36:20.440 --> 01:36:24.437
move to, you know,
unpacking more of the opportunity,

01:36:24.520 --> 01:36:26.557
uh I want to do something different. Um

01:36:26.640 --> 01:36:31.757
I want to introduce two people
whose LinkedIn uh profiles were both

01:36:31.840 --> 01:36:33.317
super impressive and slightly

01:36:33.400 --> 01:36:35.637
perplexing. Uh under current role it

01:36:35.720 --> 01:36:38.837
says they're general partners at Mantis

01:36:38.920 --> 01:36:41.557
VC. Under previous role it says they

01:36:41.640 --> 01:36:43.837
sold out Madison Square Garden. Please

01:36:43.920 --> 01:36:46.077
help me welcome Alex and Drew from The

01:36:46.160 --> 01:36:51.280
Chainsmokers. Alex, Drew. >> [music]

01:36:55.160 --> 01:36:57.197
[music] >> It's so
wonderful seeing you, Alex.

01:36:57.280 --> 01:37:01.277
Thank you so much. You know, one of the
things when my team came to me and said,

01:37:01.360 --> 01:37:04.637
"Hey, we're going to have uh
Chainsmokers at Build." I thought maybe

01:37:04.720 --> 01:37:09.757
that's what we're calling our new GitHub
Copilot app. Uh but, you know, it's so I

01:37:09.840 --> 01:37:13.877
thought maybe they're your fans.
But, tell tell me about how you got into

01:37:13.960 --> 01:37:18.317
this. I mean, you've been at it
now for what? 12-plus years

01:37:18.400 --> 01:37:21.757
uh as venture capitalists or angel
investors first and you have your firm

01:37:21.840 --> 01:37:24.877
now for 7-plus years and you even picked

01:37:24.960 --> 01:37:28.957
what I would have not thought is the
natural place, which is B2B SaaS even as

01:37:29.040 --> 01:37:32.677
the first place. We just give us
a little bit about the back story on this.

01:37:32.760 --> 01:37:36.437
>> Sure. Well, uh hey, I'm sure you guys
are wondering what timeline you're on

01:37:36.520 --> 01:37:41.577
where The Chainsmokers are at Microsoft
Build, but hey, how are how are you? Um

01:37:41.660 --> 01:37:43.117
>> [laughter] >> And uh you
know, we've been making we've

01:37:43.200 --> 01:37:45.677
been in The Chainsmokers for 14 years

01:37:45.760 --> 01:37:47.637
and um you know, when our our music

01:37:47.720 --> 01:37:50.437
started to take off in the mid-2010s, we

01:37:50.520 --> 01:37:53.477
got to play a few events like this

01:37:53.560 --> 01:37:57.997
and we met a lot of the founders
from the consumer mobile cloud era of

01:37:58.080 --> 01:38:02.557
startups and they taught us a lot and gave
us a front row seat into what early

01:38:02.640 --> 01:38:08.317
stage investing was like.
We fell in love with it. We found a lot of

01:38:08.400 --> 01:38:10.997
we had a lot in common with these
founders and the way they built their

01:38:11.080 --> 01:38:15.157
business was the same way that we
thought about starting the Chainsmokers

01:38:15.240 --> 01:38:19.157
and and you know, kind of breaking
through the noise there and you know, we

01:38:19.240 --> 01:38:22.637
got to participate in a few of their
deals. We were very lucky and we decided

01:38:22.720 --> 01:38:25.277
to institutionalize in 2020
and start our own fund.

01:38:25.360 --> 01:38:29.677
>> So it's so cool and and when you sort
of hear about all this AI stuff and what's

01:38:29.760 --> 01:38:33.037
happening. Obviously there's a lot even
going on even in your portfolio. We were

01:38:33.120 --> 01:38:35.277
talking backstage, things are changing.

01:38:35.360 --> 01:38:38.357
How do you sort of see
the opportunity going forward?

01:38:38.440 --> 01:38:40.597
>> I mean there's so
many vectors you could

01:38:40.680 --> 01:38:44.317
talk about this on. Of course there's
the creative output side which you know,

01:38:44.400 --> 01:38:47.677
we've been experimenting with in music
but always important to have your

01:38:47.760 --> 01:38:51.717
authenticity and when it comes
to creativity. But on the investment side,

01:38:51.800 --> 01:38:55.477
I think we're moving from producing
outputs to producing actions which I

01:38:55.560 --> 01:38:59.117
think presents like a very interesting
opportunity to kind of reimagine the

01:38:59.200 --> 01:39:03.397
entire architecture of the way software
enterprises been built. So instead of

01:39:03.480 --> 01:39:06.957
humans, you know, producing outputs,
it's machines producing outputs and

01:39:07.040 --> 01:39:10.957
rethinking what that entire
space looks like in that context.

01:39:11.040 --> 01:39:13.717
>> Yeah, I know it's fantastic to see that

01:39:13.800 --> 01:39:16.357
and maybe to just to close out any sort

01:39:16.440 --> 01:39:19.637
of when as artists who have had you

01:39:19.720 --> 01:39:23.757
know, great success,
when you look at a founder,

01:39:23.840 --> 01:39:28.077
what sort of is the advice you give them
when you're creating, you know, you're

01:39:28.160 --> 01:39:32.157
bringing something new and it's
a creative process. So what is

01:39:32.240 --> 01:39:35.557
sort of what are you looking for in
founders? What is it that you're giving

01:39:35.640 --> 01:39:38.397
them as advice? >> Totally.
I mean there really are so many

01:39:38.480 --> 01:39:39.997
parallels. Maybe it seems like that.

01:39:40.080 --> 01:39:43.477
Maybe it doesn't but I mean
it's really difficult as artists

01:39:43.560 --> 01:39:46.277
to really find, I guess, what your sound

01:39:46.360 --> 01:39:50.557
is, what's authentically you,
what you can continue to do over and

01:39:50.640 --> 01:39:54.717
over because with this much competition,
you have to be moving authentically, you

01:39:54.800 --> 01:39:59.157
have to be connected to the product that
you're creating, and cuz you're going to

01:39:59.240 --> 01:40:02.357
have to iterate and keep it going
and have consistency for a long period of

01:40:02.440 --> 01:40:06.077
time to kind of lock into your fan base
and make something that's special and

01:40:06.160 --> 01:40:09.917
unique. And, you know, we try to advise
our founders to do the same thing.

01:40:10.000 --> 01:40:12.077
>> Fantastic. Thank you both for joining

01:40:12.160 --> 01:40:17.037
us. Are you game for playing for us
this evening for our closest friends here?

01:40:17.120 --> 01:40:20.087
>> I will be there.
>> 6:00 6:00 this evening.

01:40:20.170 --> 01:40:22.077
>> [applause] >> Thank
you so much. Thank you.

01:40:22.160 --> 01:40:25.040
>> Thank you. Thank you.

01:40:27.000 --> 01:40:29.957
>> Back We'll be right here at 6:00 p.m.

01:40:30.040 --> 01:40:32.237
tonight, and really looking forward to

01:40:32.320 --> 01:40:36.677
it. So, so that effectively sort of

01:40:36.760 --> 01:40:38.277
talks a little bit about the developer

01:40:38.360 --> 01:40:41.077
stack. So, now let's talk about

01:40:41.160 --> 01:40:44.437
what is opportunity for every company,

01:40:44.520 --> 01:40:48.997
right? At the end of the day, we are
institution and organizational builders.

01:40:49.080 --> 01:40:52.917
How if whether it's an AI native
company, whether it's a SaaS company,

01:40:53.000 --> 01:40:55.397
whether it's an enterprise, the first

01:40:55.480 --> 01:40:58.357
thing we want to do is make sure that as

01:40:58.440 --> 01:41:01.957
you build things like plugins or agents

01:41:02.040 --> 01:41:06.757
or your AI apps, we want to make sure
that they're discovered throughout the

01:41:06.840 --> 01:41:10.557
Microsoft ecosystem.
That's job number one for us. That's

01:41:10.640 --> 01:41:12.317
what we why we're doing the things we're

01:41:12.400 --> 01:41:15.517
doing in Windows or the Microsoft 365

01:41:15.600 --> 01:41:18.357
Copilot or in Teams or in GitHub. We

01:41:18.440 --> 01:41:23.677
want to structure them such that your
applications, your agents, your plugins

01:41:23.760 --> 01:41:27.677
are discovered. Customers
are also building lots of line

01:41:27.760 --> 01:41:29.797
of business applications, line of

01:41:29.880 --> 01:41:34.717
business agents then using Copilot
Studio, and we want to make that all

01:41:34.800 --> 01:41:36.397
discoverable again as part of the

01:41:36.480 --> 01:41:38.757
Copilot experience. And Teams in some

01:41:38.840 --> 01:41:41.997
sense has become this destination

01:41:42.080 --> 01:41:44.517
for multiplayer human-to-agent

01:41:44.600 --> 01:41:47.237
interaction. We want you to be able to

01:41:47.320 --> 01:41:52.597
find agents, interact with agents
uh right in Teams. Um and we're

01:41:52.680 --> 01:41:54.237
supercharging

01:41:54.320 --> 01:41:57.517
all of this. Uh Copilot continues to

01:41:57.600 --> 01:42:00.677
evolve very quickly. It started first

01:42:00.760 --> 01:42:03.357
with chat with some of the best models,

01:42:03.440 --> 01:42:08.317
with great access to Work IQ. We didn't
have the name Work IQ, but now uh you

01:42:08.400 --> 01:42:09.877
that's kind of where it got started.

01:42:09.960 --> 01:42:12.717
Then came Co-work, uh a new way of

01:42:12.800 --> 01:42:15.557
working and generating these stunning

01:42:15.640 --> 01:42:19.077
artifacts and solve these multi-step

01:42:19.160 --> 01:42:21.117
problems, right? So, you assign a

01:42:21.200 --> 01:42:24.237
multi-step task uh to Co-work. Uh you

01:42:24.320 --> 01:42:28.357
saw GitHub.
GitHub is continuing to evolve as well.

01:42:28.440 --> 01:42:30.677
Um and now as of, you know, the come

01:42:30.760 --> 01:42:31.957
summer,

01:42:32.040 --> 01:42:35.557
uh you will be bringing coding to all

01:42:35.640 --> 01:42:39.157
knowledge work within one Copilot super

01:42:39.240 --> 01:42:44.197
app, right? That's going to be
really exciting to see. Yeah.

01:42:44.280 --> 01:42:44.917
>> [applause]

01:42:45.000 --> 01:42:48.197
>> So, you'll have chat, Co-work, and code

01:42:48.280 --> 01:42:51.517
all in Copilot. But today, we're

01:42:51.600 --> 01:42:55.837
introducing something completely new.

01:42:55.920 --> 01:43:00.597
Autopilots. Uh we can
think of Autopilots as

01:43:00.680 --> 01:43:03.557
enterprise-grade claws.

01:43:03.640 --> 01:43:07.157
These are autonomous long-running agents

01:43:07.240 --> 01:43:10.557
with full enterprise compliance that run

01:43:10.640 --> 01:43:14.077
in your tenant. Uh Autopilots can have a

01:43:14.160 --> 01:43:17.997
name, personality, custom connectors,

01:43:18.080 --> 01:43:20.957
context, and memory. And they're totally

01:43:21.040 --> 01:43:23.677
new way to reduce toil and get you back

01:43:23.760 --> 01:43:26.717
to what you love. Um to kick things off,

01:43:26.800 --> 01:43:30.037
the first Autopilot we're introducing is

01:43:30.120 --> 01:43:33.880
Scout. Let's take a look.

01:43:34.680 --> 01:43:38.700
All right. >> [applause]

01:43:39.400 --> 01:43:41.757
>> As you can see, Scout works where you

01:43:41.840 --> 01:43:47.157
work, uh joining group chats in Teams,
handling threads in Outlook, uh starting

01:43:47.240 --> 01:43:51.637
today for those of you who are on Copilot
Frontier, you can try out Scout,

01:43:51.720 --> 01:43:53.837
and in the coming months, we will build

01:43:53.920 --> 01:43:59.117
this out uh to a complete digital team
of autopilots right inside a Copilot,

01:43:59.200 --> 01:44:02.597
right? So, you can go to the Copilot app.
Scout is the one that comes by

01:44:02.680 --> 01:44:07.117
default, but you can build
more of these autopilots. Uh

01:44:07.200 --> 01:44:09.557
And so, that's the future of what we

01:44:09.640 --> 01:44:11.717
think of as the Copilot ecosystem

01:44:11.800 --> 01:44:16.797
itself. And so far, we have talked a lot
about what you can do to build your own

01:44:16.880 --> 01:44:20.597
agents and how these agents are
discovered in things like Copilot and

01:44:20.680 --> 01:44:25.277
Teams and Windows. But when
you think about what makes any

01:44:25.360 --> 01:44:27.437
organization, any enterprise, any

01:44:27.520 --> 01:44:29.797
company unique,

01:44:29.880 --> 01:44:35.037
it is its tacit knowledge that it's
continuously compounding through its

01:44:35.120 --> 01:44:39.037
operations. So, the key
consideration at some level

01:44:39.120 --> 01:44:41.437
in an AI age is to ask the question,

01:44:41.520 --> 01:44:43.357
what's the future of the firm?

01:44:43.440 --> 01:44:46.557
Uh how do you continue to preserve and

01:44:46.640 --> 01:44:50.437
compound that tacit knowledge in the age

01:44:50.520 --> 01:44:54.677
of AI where models can learn anything
from the data and the trajectories they

01:44:54.760 --> 01:44:59.317
see? Uh to do that, we believe that every

01:44:59.400 --> 01:45:02.757
organization, whether
it's an AI-native company,

01:45:02.840 --> 01:45:05.397
a SaaS company, or any enterprise, will

01:45:05.480 --> 01:45:08.797
need to build their own hill-climbing

01:45:08.880 --> 01:45:10.037
machine.

01:45:10.120 --> 01:45:13.797
It's a system that continuously improves

01:45:13.880 --> 01:45:17.237
against your objectives, your private

01:45:17.320 --> 01:45:21.717
evals, compounding
your advantage over time,

01:45:21.800 --> 01:45:23.477
not someone else's.

01:45:23.560 --> 01:45:26.117
Uh to deliver on this, we're taking the

01:45:26.200 --> 01:45:29.117
next step with frontier tuning. Uh we

01:45:29.200 --> 01:45:33.477
have been working with many customers
already to help with, you know, really

01:45:33.560 --> 01:45:37.237
build their own uh learning
loop, uh the environment,

01:45:37.320 --> 01:45:40.277
the context, the tools, the rubrics, and

01:45:40.360 --> 01:45:42.437
even train their own models.

01:45:42.520 --> 01:45:45.477
And today really we're really thrilled

01:45:45.560 --> 01:45:50.517
and excited about supercharging that
frontier tuning capability uh with the

01:45:50.600 --> 01:45:54.717
innovation that's coming out of our super
intelligence lab. To tell you all

01:45:54.800 --> 01:45:59.720
about this, let me hand it
over to Mustafa. Mustafa?

01:46:00.160 --> 01:46:01.957
>> Thank you.

01:46:02.040 --> 01:46:04.877
Thank you. Thank you.

01:46:04.960 --> 01:46:08.197
Thank you. Good morning, everybody. Um

01:46:08.280 --> 01:46:10.477
you know, we really are living in the

01:46:10.560 --> 01:46:13.677
most remarkable times. Since I started

01:46:13.760 --> 01:46:16.277
working in AI, the compute that we use

01:46:16.360 --> 01:46:19.437
to train frontier models has increased

01:46:19.520 --> 01:46:22.437
by one trillionfold.

01:46:22.520 --> 01:46:24.357
That's 12 orders of magnitude of

01:46:24.440 --> 01:46:28.317
computation in just 15 years.

01:46:28.400 --> 01:46:30.677
It's now clear that a consistent

01:46:30.760 --> 01:46:33.477
exponential increase in computation

01:46:33.560 --> 01:46:38.717
leads to predictable
advances in AI capabilities.

01:46:38.800 --> 01:46:42.757
And in the next few years, we're going
to see three more orders of magnitude of

01:46:42.840 --> 01:46:46.717
compute applied to train frontier models.

01:46:46.800 --> 01:46:48.677
Intelligence is now a function of

01:46:48.760 --> 01:46:51.397
compute. Log-linear hill climbing has

01:46:51.480 --> 01:46:54.077
become the norm. The scaling laws are

01:46:54.160 --> 01:46:56.757
clearly holding, and it is a remarkable

01:46:56.840 --> 01:46:59.637
time in our industry.

01:46:59.720 --> 01:47:03.117
And so, in this context, we at MAI are

01:47:03.200 --> 01:47:07.317
building towards what we call
humanist superintelligence.

01:47:07.400 --> 01:47:10.077
State-of-the-art AI capabilities that

01:47:10.160 --> 01:47:12.837
are explicitly designed to serve people

01:47:12.920 --> 01:47:16.677
and organizations and not replace them.

01:47:16.760 --> 01:47:19.637
Because the type of AI that we create

01:47:19.720 --> 01:47:24.477
really does matter.
We need an AI that places humanity

01:47:24.560 --> 01:47:26.037
first,

01:47:26.120 --> 01:47:28.677
that always prioritizes human [snorts]

01:47:28.760 --> 01:47:32.037
well-being and human progress.

01:47:32.120 --> 01:47:36.557
This is the core philosophy and
motivation behind our superintelligence

01:47:36.640 --> 01:47:38.957
efforts at Microsoft, and it shapes

01:47:39.040 --> 01:47:41.197
everything that we do.

01:47:41.280 --> 01:47:43.797
And as a platform company, our job and

01:47:43.880 --> 01:47:46.517
our commitment is to keep you developers

01:47:46.600 --> 01:47:49.437
building at the absolute frontier.

01:47:49.520 --> 01:47:51.837
So today, we are very excited to

01:47:51.920 --> 01:47:57.477
announce a family of seven new models
across image, voice, transcription, and

01:47:57.560 --> 01:48:01.350
coding. >> [cheering] [applause]

01:48:03.640 --> 01:48:05.677
>> These are all built with real attention

01:48:05.760 --> 01:48:07.717
to detail and a commitment to making

01:48:07.800 --> 01:48:12.757
very practical and efficient tools
that are tuned to just how you work in the

01:48:12.840 --> 01:48:14.077
real world.

01:48:14.160 --> 01:48:17.717
So first up, MAI image 2.5 and its flash

01:48:17.800 --> 01:48:20.117
variant, two super strong models that

01:48:20.200 --> 01:48:23.677
deliver a step change in quality, now at

01:48:23.760 --> 01:48:27.797
number two on the leaderboard surpassing
the score of Nana Banana 2 on image

01:48:27.880 --> 01:48:31.490
editing. >> [applause]

01:48:32.560 --> 01:48:37.357
>> They give you precise editing
with incredible control and consistency.

01:48:37.440 --> 01:48:41.997
Flash is here for super efficient
production workloads, while 2.5 gives

01:48:42.080 --> 01:48:46.237
you that maximum fidelity and
professional-grade performance. They're

01:48:46.320 --> 01:48:49.837
live in PowerPoint today, they're
rolling out to OneDrive, and right now

01:48:49.920 --> 01:48:51.757
you can access them on Foundry at a

01:48:51.840 --> 01:48:54.717
market-leading quality per dollar.

01:48:54.800 --> 01:48:59.037
Next up, we've got MAI transcribe 1.5.

01:48:59.120 --> 01:49:01.557
This is the best transcription model in

01:49:01.640 --> 01:49:04.037
the world. State-of-the-art accuracy

01:49:04.120 --> 01:49:09.437
across 43 languages, beating out Gemini
and OpenAI's flagship transcription

01:49:09.520 --> 01:49:12.797
models. We've optimized
it for real-world use,

01:49:12.880 --> 01:49:14.957
so that you can produce highly accurate

01:49:15.040 --> 01:49:18.037
transcripts for any bespoke use case

01:49:18.120 --> 01:49:22.717
five times faster than all rival models.

01:49:22.800 --> 01:49:24.237
It's now being integrated inside of

01:49:24.320 --> 01:49:27.637
GitHub Teams Copilot, Dynamics 365

01:49:27.720 --> 01:49:31.357
contact center, and it's now also
available in Foundry where I'm very

01:49:31.440 --> 01:49:36.117
excited to say it is the fastest,
most efficient, and most cost-effective

01:49:36.200 --> 01:49:40.117
transcription model of any
of the hyperscalers out there.

01:49:40.200 --> 01:49:42.880
>> [cheering] [applause]

01:49:45.000 --> 01:49:48.837
>> So, paired with that, we've got
MAI Voice 2. This is our latest speech

01:49:48.920 --> 01:49:53.557
generation model. It has beautiful
prosody, natural-sounding delivery,

01:49:53.640 --> 01:49:58.037
fine-grained emotional control, and it's
available in 15 languages with many more

01:49:58.120 --> 01:50:02.517
coming soon. We're also
announcing Voice 2 Flash, and

01:50:02.600 --> 01:50:07.117
that provides the very best value and
speed for ultra-latency sensitive voice

01:50:07.200 --> 01:50:08.997
agents, which of course is the big thing

01:50:09.080 --> 01:50:11.317
in 2026.

01:50:11.400 --> 01:50:14.197
Next up, our text foundation model,

01:50:14.280 --> 01:50:16.957
MAI Thinking 1. This is our first

01:50:17.040 --> 01:50:19.077
reasoning model, and it's exceptionally

01:50:19.160 --> 01:50:21.557
strong in our target use cases of

01:50:21.640 --> 01:50:24.917
reasoning and SWE tasks. It's a 35

01:50:25.000 --> 01:50:30.077
billion active parameter MOE with a 256K
context window. That means that it

01:50:30.160 --> 01:50:33.077
competes in the medium-sized weight
class where it's certainly punching

01:50:33.160 --> 01:50:36.997
above its weight,
and independent human raters on Surge

01:50:37.080 --> 01:50:39.037
prefer it in overall quality

01:50:39.120 --> 01:50:44.317
side-by-sides versus Sonnet 4.6.

01:50:44.400 --> 01:50:45.597
>> [applause]

01:50:45.680 --> 01:50:50.197
>> It's achieved 95
97% on AMIE 2025, which

01:50:50.280 --> 01:50:51.517
is obviously the key measure of its

01:50:51.600 --> 01:50:53.837
general-purpose reasoning abilities, but

01:50:53.920 --> 01:50:57.837
most importantly of all, it's now at 53%

01:50:57.920 --> 01:51:00.197
on SWE Bench Pro, which places it right

01:51:00.280 --> 01:51:03.077
alongside Opus 4.6, at least on the

01:51:03.160 --> 01:51:07.897
toughest coding benchmark that's out
there. So, we're very happy with that.

01:51:07.980 --> 01:51:12.357
>> [cheering and applause] >> Now,
there's plenty more for us to do as

01:51:12.440 --> 01:51:16.477
we get this into production and hill
climb against real world uh tasks and

01:51:16.560 --> 01:51:19.557
real world traffic. But,
it's what is actually most

01:51:19.640 --> 01:51:24.597
remarkable about this model we think
is that it has climbed entirely from the

01:51:24.680 --> 01:51:27.517
bottom. And that means
that it hasn't targeted

01:51:27.600 --> 01:51:31.877
any of the benchmarks specifically,
and it's done so with absolutely zero

01:51:31.960 --> 01:51:36.117
distillation. And to us, this is critical.
Because it means that the

01:51:36.200 --> 01:51:37.917
model is created with an

01:51:38.000 --> 01:51:43.157
enterprise-grade, clean, and commercially
licensed data lineage. That

01:51:43.240 --> 01:51:46.677
means that you can put it into
production in a very trustworthy way

01:51:46.760 --> 01:51:49.117
with complete confidence.

01:51:49.200 --> 01:51:51.157
Now, finally, I'm incredibly excited to

01:51:51.240 --> 01:51:56.157
announce MAI Code 1 Flash.
This is our new inference efficient

01:51:56.240 --> 01:52:00.077
coding model, which has been especially
tuned for VS Code and of course GitHub

01:52:00.160 --> 01:52:01.957
Copilot CLI.

01:52:02.040 --> 01:52:05.397
It achieves 51% on SWE-Bench Pro despite

01:52:05.480 --> 01:52:08.197
having just 5 billion parameters.

01:52:08.280 --> 01:52:13.277
And so, it's much closer to I Q in terms
of size, but cheaper in cost, delivering

01:52:13.360 --> 01:52:15.197
really strong coding performance at

01:52:15.280 --> 01:52:21.257
great inference efficiency. And it's
rolling out today inside of VS Code.

01:52:21.340 --> 01:52:23.360
>> [applause]

01:52:25.080 --> 01:52:29.157
>> Now, alongside distribution on Foundry
and optimization for our 1P products,

01:52:29.240 --> 01:52:33.757
we're also very excited to make our
models available on Open Router as well

01:52:33.840 --> 01:52:37.117
as Fireworks and Base Ten. This means
that for the first time, you're going to

01:52:37.200 --> 01:52:38.797
be able to tune the weights directly

01:52:38.880 --> 01:52:42.437
yourself in an ecosystem of your choice.

01:52:42.520 --> 01:52:44.997
Now, across this entire family, safety

01:52:45.080 --> 01:52:49.757
and security have been built in from
the start. Our voice models come with

01:52:49.840 --> 01:52:53.757
protections against unauthorized cloning.
Everything is watermarked from

01:52:53.840 --> 01:52:58.117
scratch. We've reduced our over
refusals, improved representation,

01:52:58.200 --> 01:53:00.437
including for people with disabilities.

01:53:00.520 --> 01:53:02.797
We're also publishing a very detailed

01:53:02.880 --> 01:53:04.997
technical report today to give you to

01:53:05.080 --> 01:53:10.397
full and transparent understanding
of how we put all of this together.

01:53:10.480 --> 01:53:13.957
Now, one of the things I'm particularly
excited about is that we have been

01:53:14.040 --> 01:53:16.557
carefully co-designing our models with

01:53:16.640 --> 01:53:18.837
our own silicon. So, that means that

01:53:18.920 --> 01:53:21.517
we've optimized MAI thinking one on our

01:53:21.600 --> 01:53:24.597
very own Maia 200 chip and benchmarked

01:53:24.680 --> 01:53:28.157
it head-to-head against the GB200.

01:53:28.240 --> 01:53:30.757
And so, on top of the 30% performance

01:53:30.840 --> 01:53:32.517
improvement that Satya talked about

01:53:32.600 --> 01:53:36.477
earlier, we're now seeing a further 1.4x

01:53:36.560 --> 01:53:38.877
performance per watt gain when we run

01:53:38.960 --> 01:53:41.997
our MAI models on the Maia 200

01:53:42.080 --> 01:53:44.557
end-to-end. And that's huge. Cuz as

01:53:44.640 --> 01:53:47.277
everybody knows, at this scale, every

01:53:47.360 --> 01:53:51.717
watt counts. And silicon
and model co-design is a

01:53:51.800 --> 01:53:56.077
really key advantage that we think
is going to help keep everybody here right

01:53:56.160 --> 01:54:00.117
on the frontier with the most efficient
and most powerful thinking and coding

01:54:00.200 --> 01:54:04.077
agents out there.
We're also super excited that these

01:54:04.160 --> 01:54:06.317
faster and more efficient MAI models are

01:54:06.400 --> 01:54:11.157
coming to the N1X that Satya mentioned
a few moments ago. And we think that's

01:54:11.240 --> 01:54:15.197
going to be able to deliver the very
best performance uh on Windows in a few

01:54:15.280 --> 01:54:16.997
months' time.

01:54:17.080 --> 01:54:19.797
Now, to us, this is what owning the full

01:54:19.880 --> 01:54:22.237
stack end-to-end looks like. It's the

01:54:22.320 --> 01:54:25.277
foundation of Microsoft frontier tuning,

01:54:25.360 --> 01:54:29.437
lets you customize the MAI models using
our full stack hill climbing machine

01:54:29.520 --> 01:54:33.797
right where you want it. And it means
that the disciplined and very relentless

01:54:33.880 --> 01:54:38.197
engineering that is going into building
our models is now available to all of

01:54:38.280 --> 01:54:40.637
you on a platform that you can trust,

01:54:40.720 --> 01:54:43.397
working on your behalf to create custom

01:54:43.480 --> 01:54:48.477
agents that you will control.
So, the really big thing, of course,

01:54:48.560 --> 01:54:51.597
that's happened in the last year
is these ARELEs, reinforcement learning

01:54:51.680 --> 01:54:55.877
environments, these unique
training gyms for your AIs.

01:54:55.960 --> 01:54:58.397
They create company and task-specific

01:54:58.480 --> 01:55:02.997
agents adapted only to you
built on our MAI models.

01:55:03.080 --> 01:55:07.197
So, for example, within Microsoft,
we use our RLEs combined with our MAI

01:55:07.280 --> 01:55:09.637
models to climb towards the best agentic

01:55:09.720 --> 01:55:11.957
use cases on Excel.

01:55:12.040 --> 01:55:14.317
Our MAI tuned model is now on par with

01:55:14.400 --> 01:55:16.957
GPT-5.4 on public and private

01:55:17.040 --> 01:55:19.277
benchmarks, whilst at the same time

01:55:19.360 --> 01:55:24.237
being 10 times more efficient on cost.

01:55:24.320 --> 01:55:29.757
Thank you. You know, and many other
early adopters are seeing similar results.

01:55:29.840 --> 01:55:32.837
When we've tuned our models
on McKinsey's tasks, MAI delivered the

01:55:32.920 --> 01:55:36.837
highest win rate,
even outperforming GPT-5.5,

01:55:36.920 --> 01:55:39.437
and again delivering 10x greater

01:55:39.520 --> 01:55:41.317
efficiency on cost.

01:55:41.400 --> 01:55:44.117
So, to us, this is the advantage of very

01:55:44.200 --> 01:55:47.677
carefully calibrated frontier tuning.

01:55:47.760 --> 01:55:51.437
And importantly, unlike with
some of the other companies,

01:55:51.520 --> 01:55:56.157
with MAI, you don't rent intelligence
from a shared model that learns from

01:55:56.240 --> 01:55:59.557
everybody. Only you
keep the benefits of your

01:55:59.640 --> 01:56:05.317
hard-earned workflows, know-how,
knowledge, and your own institutional

01:56:05.400 --> 01:56:08.397
data. Only you get
to control the resulting

01:56:08.480 --> 01:56:11.677
model. And so, with us, the RLEs and the

01:56:11.760 --> 01:56:16.117
models that you build inside
of them, they become your moat.

01:56:16.200 --> 01:56:20.717
I really think this is distinct.
It marks a new era in AI that we're all

01:56:20.800 --> 01:56:24.117
very, very excited about. Okay.

01:56:24.200 --> 01:56:26.557
So, now just one final announcement that

01:56:26.640 --> 01:56:28.797
I'm very excited about. We are taking

01:56:28.880 --> 01:56:33.797
customization and co-creation of our
models to the highest level possible on

01:56:33.880 --> 01:56:35.717
what I think of as perhaps the most

01:56:35.800 --> 01:56:39.477
important application of AI, healthcare.

01:56:39.560 --> 01:56:42.557
So, today, we're very proud to be
announcing that we're partnering with

01:56:42.640 --> 01:56:46.957
Mayo Clinic to jointly develop a new
frontier model for health, and then

01:56:47.040 --> 01:56:50.797
deploy it around the world in their
hospitals and beyond. So, this morning,

01:56:50.880 --> 01:56:52.917
please help me in welcoming to the stage

01:56:53.000 --> 01:56:54.957
a physician, groundbreaking researcher,

01:56:55.040 --> 01:57:01.960
president, and CEO of the Mayo
Clinic, Dr. Gianrico Farrugia.

01:57:02.340 --> 01:57:05.010
>> [applause] [music]

01:57:06.320 --> 01:57:08.717
>> Thank you so much
for being here, Gianrico.

01:57:08.800 --> 01:57:13.197
Um now, of course, everyone will
recognize Mayo as perhaps the leading

01:57:13.280 --> 01:57:17.917
hospital in the world with an incredible
track record of research and innovation

01:57:18.000 --> 01:57:21.757
and clinical practice. Tell us a little
bit more about what you hope to get out

01:57:21.840 --> 01:57:23.997
of our collaboration. >> Well,
first of all, thanks for having me

01:57:24.080 --> 01:57:28.437
here. Thanks to Statis Mayo
Clinic is known for being able to

01:57:28.520 --> 01:57:33.077
live up to our primary value. The needs
of the patient come first. We deliver

01:57:33.160 --> 01:57:37.157
outstanding health care. We're ranked
the number one health care organization

01:57:37.240 --> 01:57:40.797
in the world. Yet,
we know most people in the world

01:57:40.880 --> 01:57:43.117
will not have access to Mayo Clinic.

01:57:43.200 --> 01:57:47.717
So, 7 years ago, we decided to create
a platform, the Mayo Clinic platform,

01:57:47.800 --> 01:57:49.277
moving all of health care from a

01:57:49.360 --> 01:57:51.837
pipeline to a platform.

01:57:51.920 --> 01:57:54.837
And with our partners, that platform now

01:57:54.920 --> 01:57:57.317
is in four continents and reaches about

01:57:57.400 --> 01:58:02.157
100 million people.
It has created the largest, to our

01:58:02.240 --> 01:58:06.277
knowledge, deepest
longitudinal health care data

01:58:06.360 --> 01:58:10.957
set in the world, multimodal,
gen- including genomics. So, here together

01:58:11.040 --> 01:58:14.677
now we have the opportunity to do what
we do best together, which is create a

01:58:14.760 --> 01:58:16.877
frontier model for health care.

01:58:16.960 --> 01:58:20.837
What it means if you're a patient,
if you're somebody interested in health

01:58:20.920 --> 01:58:25.437
care, you can get clinical and logistical
answers to your health care.

01:58:25.520 --> 01:58:28.197
But, if you're health care
provider, if you're a physician,

01:58:28.280 --> 01:58:30.837
it can give you insights, can act as

01:58:30.920 --> 01:58:33.797
your real-time team member that can tell

01:58:33.880 --> 01:58:37.397
you what is likely to happen next, but

01:58:37.480 --> 01:58:39.277
it can also prevent harm and therefore

01:58:39.360 --> 01:58:44.477
increase patient safety, and giving
valuable insights that make the team

01:58:44.560 --> 01:58:47.877
better at giving you what you need
most, which is better health care.

01:58:47.960 --> 01:58:49.237
>> Yeah, and I think
that one of the things

01:58:49.320 --> 01:58:52.877
that we're most excited about is that
the models are already pretty incredible

01:58:52.960 --> 01:58:57.157
at textbook knowledge. It read all
the journals and all the papers, but what

01:58:57.240 --> 01:59:01.077
they're really lacking is the kind of
clinical practice and clinical expertise

01:59:01.160 --> 01:59:04.517
of your team and your clinicians
that you've developed over the last many

01:59:04.600 --> 01:59:08.877
decades. So, how do you think we might
go about using that clinical practice to

01:59:08.960 --> 01:59:11.317
improve the performance
of the model in production?

01:59:11.400 --> 01:59:15.797
>> So, the exciting part here is we each
do what we do best, and we can tackle

01:59:15.880 --> 01:59:17.357
something that has eluded health care

01:59:17.440 --> 01:59:20.197
for a long time, trusted, scalable

01:59:20.280 --> 01:59:25.037
solutions. And to do that, you need
to have the right data, you certainly need

01:59:25.120 --> 01:59:28.917
to have the right people, but you also
need to have a very patient-focused

01:59:29.000 --> 01:59:30.957
lens. And between the two of us, we now

01:59:31.040 --> 01:59:34.837
have these all this together, so we can

01:59:34.920 --> 01:59:39.397
build this frontier model, and then
offer safe, secure, trustworthy, and of

01:59:39.480 --> 01:59:42.557
course effective health
care solutions for all.

01:59:42.640 --> 01:59:45.277
>> Well, our number one objective,
of course, is to put the patient first,

01:59:45.360 --> 01:59:48.717
deliver the highest quality we can
in a trusted way, and then hopefully share

01:59:48.800 --> 01:59:52.157
that with the world over. So, we're very
excited for this partnership. Can't wait

01:59:52.240 --> 01:59:53.717
to share more with everyone in the future.

01:59:53.800 --> 01:59:57.200
>> Us too. Thank you. >> Thank you.

01:59:57.880 --> 01:59:59.900
>> [applause]

02:00:01.240 --> 02:00:04.397
>> So, today marks
some very, very exciting

02:00:04.480 --> 02:00:07.917
steps that we're taking on our journey
to create humanistic superintelligence

02:00:08.000 --> 02:00:12.877
at Microsoft. We now have an incredible
roster of seven new world-class models

02:00:12.960 --> 02:00:15.997
to keep everybody working
at the absolute frontier.

02:00:16.080 --> 02:00:19.917
And we're really looking forward
to everybody being able to co-create your

02:00:20.000 --> 02:00:22.597
own unique agents, adapted to you, that

02:00:22.680 --> 02:00:24.837
you'll control. I really feel like this

02:00:24.920 --> 02:00:28.717
is a new era in AI, an era of AI that

02:00:28.800 --> 02:00:33.077
you control on your terms. So, let's
build it together. Thank you very much,

02:00:33.160 --> 02:00:36.580
everyone. >> [applause]

02:00:40.440 --> 02:00:46.600
>> And now over to Tanya to show us
how frontier tuning works in practice.

02:00:47.660 --> 02:00:49.680
>> [applause]

02:00:52.520 --> 02:00:56.757
>> Thanks, Mustafa.
With frontier tuning, we're making it

02:00:56.840 --> 02:00:59.397
possible for you to create your own

02:00:59.480 --> 02:01:01.237
enterprise AI

02:01:01.320 --> 02:01:04.717
with the models and the harness tuned on

02:01:04.800 --> 02:01:07.037
your data and your workflows. Let me

02:01:07.120 --> 02:01:12.080
show you how you can build
your own hill climbing machine.

02:01:12.400 --> 02:01:17.277
MAI thinking one is now available
in private preview in the foundry model

02:01:17.360 --> 02:01:21.077
catalog. You can go ahead
and deploy the model as

02:01:21.160 --> 02:01:23.957
is, or if you want to go on your own

02:01:24.040 --> 02:01:26.237
hill climbing journey, you can start by

02:01:26.320 --> 02:01:29.237
clicking the fine-tune button.

02:01:29.320 --> 02:01:31.437
Now, I'm in the fine-tuning UI. The

02:01:31.520 --> 02:01:34.677
first thing I'll do is add my data set.

02:01:34.760 --> 02:01:39.157
Next, I'm going to add a grader,
and that's it. I go ahead and submit the

02:01:39.240 --> 02:01:40.557
job.

02:01:40.640 --> 02:01:43.757
Couple of hours in, I see how the model

02:01:43.840 --> 02:01:46.557
is learning. It generates rollouts, it

02:01:46.640 --> 02:01:50.317
scores them, and we start to hill climb.

02:01:50.400 --> 02:01:54.557
Now, if you want full control over
your RL training loop, let me give you a

02:01:54.640 --> 02:01:56.917
sneak peek into our low-level training

02:01:57.000 --> 02:02:01.717
API. You can see here
I have the MAI thinking

02:02:01.800 --> 02:02:02.917
model.

02:02:03.000 --> 02:02:06.197
I can also configure my rollout strategy

02:02:06.280 --> 02:02:09.237
as well as my hyper parameters to define

02:02:09.320 --> 02:02:11.917
exactly how I want my training algorithm

02:02:12.000 --> 02:02:13.597
to work.

02:02:13.680 --> 02:02:16.437
In fact, I can also incorporate my own

02:02:16.520 --> 02:02:19.357
RL gym here by defining the tools that

02:02:19.440 --> 02:02:21.877
this model interacts with.

02:02:21.960 --> 02:02:24.517
All right. You just saw all the code.

02:02:24.600 --> 02:02:27.437
But as an M365 customer, you're never

02:02:27.520 --> 02:02:30.277
starting from scratch. Let's hop on to

02:02:30.360 --> 02:02:32.517
Copilot.

02:02:32.600 --> 02:02:34.917
As a part of frontier tuning, we're

02:02:35.000 --> 02:02:38.077
introducing a new way to build RL

02:02:38.160 --> 02:02:41.637
environments based on your data and your

02:02:41.720 --> 02:02:44.717
workflows. One of our customers, Land

02:02:44.800 --> 02:02:46.957
O'Lakes, one amongst the largest

02:02:47.040 --> 02:02:50.037
agribusinesses in the United States, is

02:02:50.120 --> 02:02:52.837
in fact using this to perfect that

02:02:52.920 --> 02:02:55.157
butter on your morning toast. Let me

02:02:55.240 --> 02:02:58.797
walk you through their environment.

02:02:58.880 --> 02:03:01.197
Now, on a high level, the environment

02:03:01.280 --> 02:03:03.877
consists of skills, knowledge, and

02:03:03.960 --> 02:03:07.557
tools. But, in the back end, we create

02:03:07.640 --> 02:03:12.597
an entire RL gym for your agents
to continuously learn the way that you

02:03:12.680 --> 02:03:15.277
work. Now, let's look at the butter report

02:03:15.360 --> 02:03:17.437
generation skill here.

02:03:17.520 --> 02:03:20.277
These tasks are very complex. They

02:03:20.360 --> 02:03:23.237
require many manual steps and a high

02:03:23.320 --> 02:03:28.317
degree of precision. Now,
in these tasks, even 80% accuracy

02:03:28.400 --> 02:03:30.077
isn't good enough.

02:03:30.160 --> 02:03:32.477
To help clients to higher accuracy,

02:03:32.560 --> 02:03:35.517
we're extending the industry definition

02:03:35.600 --> 02:03:38.557
of skills to include rubrics on what

02:03:38.640 --> 02:03:40.277
good looks like.

02:03:40.360 --> 02:03:42.677
But, this is just one task. How do I

02:03:42.760 --> 02:03:46.237
scale this to codify all of the tasks in

02:03:46.320 --> 02:03:50.637
enterprise? You spend
a lot of time in M365, in

02:03:50.720 --> 02:03:53.037
Teams, Outlook, Word, Excel, and

02:03:53.120 --> 02:03:56.157
PowerPoint. We use those signals to

02:03:56.240 --> 02:03:58.757
suggest skills and rubrics that truly

02:03:58.840 --> 02:04:04.037
define the way that you work.
Next, you can add your organizational

02:04:04.120 --> 02:04:07.757
knowledge for grounding,
like from OneDrive, SharePoint.

02:04:07.840 --> 02:04:12.437
Now, the environment comes built-in with
Microsoft tools, and you can add custom

02:04:12.520 --> 02:04:16.677
tools to it.
Because these tools tap into real

02:04:16.760 --> 02:04:19.077
workflows, we virtualize them to

02:04:19.160 --> 02:04:21.757
simulate execution, so the model can

02:04:21.840 --> 02:04:24.237
learn without actually impacting the

02:04:24.320 --> 02:04:28.917
live state of your business.
And now, my most favorite part, the

02:04:29.000 --> 02:04:30.557
science.

02:04:30.640 --> 02:04:33.477
By generalizing all of these learnings

02:04:33.560 --> 02:04:36.237
into the main model, as well as in the

02:04:36.320 --> 02:04:39.037
embedding model, we're able to hill

02:04:39.120 --> 02:04:44.560
climb for tasks that
require high accuracy.

02:04:45.000 --> 02:04:49.317
And not only this, we are able to hill

02:04:49.400 --> 02:04:52.797
climb to more than 90% accuracy for Land

02:04:52.880 --> 02:04:56.557
O'Lakes tasks using the MEI model.

02:04:56.640 --> 02:04:59.997
In fact, we estimate this model to be

02:05:00.080 --> 02:05:04.237
10x more efficient than
the baseline models.

02:05:04.320 --> 02:05:06.557
Now with this environment setup, let's

02:05:06.640 --> 02:05:09.357
go back to butter reporting.

02:05:09.440 --> 02:05:13.157
Let's run this task this time using this

02:05:13.240 --> 02:05:16.717
tuned environment
as an inferencing harness.

02:05:16.800 --> 02:05:19.357
Now usually this task takes a couple of

02:05:19.440 --> 02:05:21.957
minutes to run. In the meantime, I will

02:05:22.040 --> 02:05:25.957
show you a cached response.

02:05:26.040 --> 02:05:28.877
Now as you can see here, the agent has

02:05:28.960 --> 02:05:31.917
leveraged test time free search with

02:05:32.000 --> 02:05:35.197
multiple models including a fine-tuned

02:05:35.280 --> 02:05:38.237
model. And what you see here is a

02:05:38.320 --> 02:05:41.197
summary that doesn't feel generic. It

02:05:41.280 --> 02:05:44.237
feels undoubtedly Land O'Lakes. And

02:05:44.320 --> 02:05:47.357
that's not it. The task holds itself to

02:05:47.440 --> 02:05:50.157
high standards and it continuously

02:05:50.240 --> 02:05:53.637
retrospects and evaluates itself.

02:05:53.720 --> 02:05:56.877
This way, with frontier tuning, your

02:05:56.960 --> 02:05:59.877
agent continuously improves with your

02:05:59.960 --> 02:06:04.077
compliant RL environment, on your data,

02:06:04.160 --> 02:06:07.837
encoding the way you work. Now that is

02:06:07.920 --> 02:06:11.477
what I call frontier
tuning as smooth as butter.

02:06:11.560 --> 02:06:16.077
We cannot wait to see
the environments you built.

02:06:16.160 --> 02:06:19.007
Back to you, Satya. >> All right.

02:06:19.090 --> 02:06:21.077
>> [applause]

02:06:21.160 --> 02:06:24.357
>> Thank you so much, Tanya, Mustafa, Jen

02:06:24.440 --> 02:06:26.157
Rico. Thank you for the partnership

02:06:26.240 --> 02:06:29.357
here. Uh you know, what you just saw is

02:06:29.440 --> 02:06:33.197
a pretty significant shift. Uh

02:06:33.280 --> 02:06:36.317
we believe that times come for every

02:06:36.400 --> 02:06:40.077
company to just move from consuming

02:06:40.160 --> 02:06:44.237
a frontier model to fully participating

02:06:44.320 --> 02:06:45.757
at the frontier in the frontier

02:06:45.840 --> 02:06:48.917
ecosystem. That's the transition. Uh you

02:06:49.000 --> 02:06:51.917
can have your own private Evals and

02:06:52.000 --> 02:06:53.357
outcomes,

02:06:53.440 --> 02:06:56.717
your private RLs and traces, your

02:06:56.800 --> 02:06:59.077
enterprise knowledge, create the

02:06:59.160 --> 02:07:02.597
scaffolding for models to hill climb.

02:07:02.680 --> 02:07:07.917
That's what will allow you to create
that differentiated IP that you own, you

02:07:08.000 --> 02:07:09.397
control.

02:07:09.480 --> 02:07:14.997
But the second salient point is that
there is a new operating point at the

02:07:15.080 --> 02:07:16.317
frontier.

02:07:16.400 --> 02:07:20.237
Uh where you can use a very efficient uh

02:07:20.320 --> 02:07:24.517
reasoning model and a coding model and

02:07:24.600 --> 02:07:28.397
achieve frontier level performance
because you've done the hard work of

02:07:28.480 --> 02:07:30.557
creating that environment, that RL, that

02:07:30.640 --> 02:07:35.757
hill climbing machine in which these
models with your traces can hill climb

02:07:35.840 --> 02:07:39.517
to the frontier.
So we think that these combination of

02:07:39.600 --> 02:07:44.557
these two is a pretty big game-changer
on how people think about what does it

02:07:44.640 --> 02:07:48.757
mean to operate at the frontier? What
does frontier tokens look like? How are

02:07:48.840 --> 02:07:50.837
you in control? What's the future of a

02:07:50.920 --> 02:07:53.397
firm? These are the big questions and

02:07:53.480 --> 02:07:55.957
really an ecosystem that that gets built

02:07:56.040 --> 02:07:58.437
around as opposed to a few models that

02:07:58.520 --> 02:08:01.077
just are hungry for all data.

02:08:01.160 --> 02:08:03.797
To close out though, I want to talk

02:08:03.880 --> 02:08:07.717
about that frontiers beyond um and that

02:08:07.800 --> 02:08:10.997
push this ecosystem of ours and really

02:08:11.080 --> 02:08:13.437
say what's the next big thing.

02:08:13.520 --> 02:08:18.437
Bending the curve, you know, when
Mustafa talked about this with Jean Rico

02:08:18.520 --> 02:08:22.317
um when it comes to health,
but building, you know, that scientific

02:08:22.400 --> 02:08:24.397
discovery loop

02:08:24.480 --> 02:08:28.037
can have perhaps the biggest societal uh

02:08:28.120 --> 02:08:31.277
impact, right? Science today still

02:08:31.360 --> 02:08:33.317
is a little too linear. You create a

02:08:33.400 --> 02:08:36.157
hypothesis, you form a hypothesis,

02:08:36.240 --> 02:08:40.837
you run an experiment, you wait
for lab results, which is out of band,

02:08:40.920 --> 02:08:45.277
and then you begin again. But what
if the scientific method itself could

02:08:45.360 --> 02:08:49.517
become more continuous,
more parallel, more programmable?

02:08:49.600 --> 02:08:53.277
That's what we're doing with Microsoft
Discovery, right? Discovery brings

02:08:53.360 --> 02:08:57.557
together the models, the HPC compute,
the knowledge graphs of all the

02:08:57.640 --> 02:08:59.357
scientific knowledge,

02:08:59.440 --> 02:09:02.757
the automated lab and simulation into

02:09:02.840 --> 02:09:05.637
one agentic discovery loop. And so

02:09:05.720 --> 02:09:08.637
today, I'm really thrilled to GA

02:09:08.720 --> 02:09:11.397
Microsoft Discovery and and to show you

02:09:11.480 --> 02:09:15.857
all of this in action, let me invite up
on David on stage. David, take it away.

02:09:15.940 --> 02:09:19.117
>> [applause and cheering] >> Hello.

02:09:19.200 --> 02:09:22.557
I work in the Microsoft
Discovery and Quantum team.

02:09:22.640 --> 02:09:24.877
Today, to recycle a PET plastic like

02:09:24.960 --> 02:09:27.397
this bottle, you have to shred it and

02:09:27.480 --> 02:09:32.357
melt it. The result is degraded. You
cannot use it to make this bottle again.

02:09:32.440 --> 02:09:35.757
It's downcycling.
Cambridge Consultants, part of

02:09:35.840 --> 02:09:40.317
Capgemini, is using Microsoft Discovery
to advance that research. Let me show it

02:09:40.400 --> 02:09:44.917
to you. This is the Microsoft
Discovery app. It

02:09:45.000 --> 02:09:47.117
is based on VS code because agentic

02:09:47.200 --> 02:09:49.677
discovery has many parallels with

02:09:49.760 --> 02:09:54.797
agentic software engineering. Today,
I want to do these three things as a

02:09:54.880 --> 02:09:58.357
scientist. First,
I want to write a scientific

02:09:58.440 --> 02:10:00.797
paper about this topic. I'll explore an

02:10:00.880 --> 02:10:03.197
existing line of research. Instead of

02:10:03.280 --> 02:10:06.797
melting the plastic, I can use proteins

02:10:06.880 --> 02:10:11.637
to decompose it, so you can
recycle it again and again.

02:10:11.720 --> 02:10:13.277
Second,

02:10:13.360 --> 02:10:15.997
I want to perform the actual discovery

02:10:16.080 --> 02:10:19.717
to find new proteins. And third,

02:10:19.800 --> 02:10:24.957
I want to create a lab protocol
to test the results in a real lab.

02:10:25.040 --> 02:10:26.717
If you are a developer, these three

02:10:26.800 --> 02:10:29.517
steps should be very familiar to you.

02:10:29.600 --> 02:10:33.837
They look a lot like planning,
coding and testing, and deploying to

02:10:33.920 --> 02:10:36.277
production, but for science.

02:10:36.360 --> 02:10:40.917
Let's launch it. This will
kick off the discovery engine.

02:10:41.000 --> 02:10:45.917
You can think of this as a as a team of
specialized agents always running in the

02:10:46.000 --> 02:10:50.917
background following the scientific
method. You can see those agents here,

02:10:51.000 --> 02:10:52.597
and you can add more.

02:10:52.680 --> 02:10:54.997
Microsoft Discovery includes a community

02:10:55.080 --> 02:10:58.437
of agents, models, and tools across many

02:10:58.520 --> 02:11:02.317
domains. You can use open
source, third parties,

02:11:02.400 --> 02:11:06.797
or create your own. Okay,
so this is still running, and it

02:11:06.880 --> 02:11:09.197
will be running for a while, even hours

02:11:09.280 --> 02:11:11.717
or days, because the discovery engine,

02:11:11.800 --> 02:11:14.957
like science, is not sequential. It is

02:11:15.040 --> 02:11:17.477
exploring hypotheses and performing

02:11:17.560 --> 02:11:20.637
long-running simulations dynamically.

02:11:20.720 --> 02:11:24.597
So, let me open one that is already
completed. You can see the files that it

02:11:24.680 --> 02:11:29.277
created here. This one is the
research paper that I was asking

02:11:29.360 --> 02:11:33.197
for. To create it, Discovery's using a

02:11:33.280 --> 02:11:37.837
knowledge graph internally, bringing
together public scientific literature

02:11:37.920 --> 02:11:41.957
and internal knowledge.
This is a critical asset because it

02:11:42.040 --> 02:11:46.957
provides complete visibility of the
reasoning, so scientists can be in full

02:11:47.040 --> 02:11:50.517
control. We can go now
to step two, the discovery

02:11:50.600 --> 02:11:54.237
itself. Let me open
the output for that one.

02:11:54.320 --> 02:11:58.117
So, how do you come
up with new proteins to

02:11:58.200 --> 02:12:00.597
decompose plastic? This is the approach

02:12:00.680 --> 02:12:02.237
that Discovery took.

02:12:02.320 --> 02:12:05.277
First, you need an AI model to predict

02:12:05.360 --> 02:12:07.677
how good a protein is for that.

02:12:07.760 --> 02:12:09.997
Then, you need a way to generate new

02:12:10.080 --> 02:12:12.437
proteins, and finally, you need to

02:12:12.520 --> 02:12:15.637
identify the most promising ones. So,

02:12:15.720 --> 02:12:20.397
let's start with the model. Microsoft
Discovery trained multiple models and

02:12:20.480 --> 02:12:24.477
picked the best. And there's no
out-of-the-box agent for that. But,

02:12:24.560 --> 02:12:27.637
that's okay. If Microsoft
Discovery does not find an

02:12:27.720 --> 02:12:30.397
agent for a task, it will create one on

02:12:30.480 --> 02:12:32.997
the fly, which is pretty cool. So,

02:12:33.080 --> 02:12:36.197
here's all the files that it created.

02:12:36.280 --> 02:12:38.917
This one here is the YAML for the agent.

02:12:39.000 --> 02:12:43.157
And this one, for example,
is the Python code to train the models.

02:12:43.240 --> 02:12:48.117
Now, we need to generate candidates
for new proteins. And that task requires a

02:12:48.200 --> 02:12:53.037
lot of compute. Discovery
is integrated with HPC. So,

02:12:53.120 --> 02:12:56.397
agents can use it for complex simulations.

02:12:56.480 --> 02:12:58.557
You can see the process here.

02:12:58.640 --> 02:13:01.437
It started with a seed protein, and then

02:13:01.520 --> 02:13:05.957
it created variations by replacing small
segments. You can see those segments

02:13:06.040 --> 02:13:09.277
there. Then, it applied
the model from before

02:13:09.360 --> 02:13:13.797
to see the variation helps or not,
learning in that process.

02:13:13.880 --> 02:13:16.277
This is done millions of times in

02:13:16.360 --> 02:13:18.677
multiple jobs in parallel, exploring a

02:13:18.760 --> 02:13:23.317
huge tree of proteins. The result is 80

02:13:23.400 --> 02:13:25.837
proteins that are ready to be sent to

02:13:25.920 --> 02:13:30.877
the lab for testing, which we said it
was like deploying software, right? The

02:13:30.960 --> 02:13:34.517
problem, though, is that creating
a protein is a little bit more

02:13:34.600 --> 02:13:37.077
complicated. The most common way is

02:13:37.160 --> 02:13:39.917
inserting DNA into bacteria, so the

02:13:40.000 --> 02:13:42.917
bacteria create the protein for me. So,

02:13:43.000 --> 02:13:45.877
Discovery created another file.

02:13:45.960 --> 02:13:48.437
This one here. This file contains all

02:13:48.520 --> 02:13:50.477
the DNA sequences to create each

02:13:50.560 --> 02:13:53.277
protein. I can send this now to a lab,

02:13:53.360 --> 02:13:55.717
create the proteins, and test them.

02:13:55.800 --> 02:13:59.317
Or, I can go one step further.

02:13:59.400 --> 02:14:04.197
If I have an automated lab, I can
integrate it directly with the agents.

02:14:04.280 --> 02:14:07.477
Let's do it. So, let me
go back to the session,

02:14:07.560 --> 02:14:10.837
and let's just submit job to lab.

02:14:10.920 --> 02:14:14.317
Go. This will use a agent that is sending

02:14:14.400 --> 02:14:16.757
instructions for the lab equipment. And

02:14:16.840 --> 02:14:22.317
this is very real. Cambridge Consultants
does have an automated lab, and I have

02:14:22.400 --> 02:14:26.277
it right here. So,
this is the application controlling

02:14:26.360 --> 02:14:29.037
the lab. It has a Copilot interface, so

02:14:29.120 --> 02:14:33.357
the scientist can interact with the lab
to design experiments, which is It feels

02:14:33.440 --> 02:14:35.957
like being Iron Man, but for chemistry.

02:14:36.040 --> 02:14:40.277
In this case, Discovery is submitting
the job. You can see it already here.

02:14:40.360 --> 02:14:43.837
But, remember, you need to grow
the bacteria, so this will take some time.

02:14:43.920 --> 02:14:46.317
So, let me open a previous run.

02:14:46.400 --> 02:14:48.357
And there you go. These are all the

02:14:48.440 --> 02:14:51.277
steps that the agents are doing. Most of

02:14:51.360 --> 02:14:58.160
them completely automated with human
supervision. How cool is that?

02:14:58.510 --> 02:15:00.530
>> [applause]

02:15:02.880 --> 02:15:06.437
>> This is bringing together the physical
world and the digital agents in a

02:15:06.520 --> 02:15:11.157
unified discovery loop. And this was
just one example of what Microsoft

02:15:11.240 --> 02:15:13.317
Discovery can do. Customers across

02:15:13.400 --> 02:15:15.837
industries are using it today to embrace

02:15:15.920 --> 02:15:20.697
a new era of scientific discovery.
Thank you. Back to you, Satya.

02:15:20.780 --> 02:15:22.800
>> [applause]

02:15:24.080 --> 02:15:25.877
>> Thank you so much, David. You know,

02:15:25.960 --> 02:15:29.277
talking about uh scientific discovery,

02:15:29.360 --> 02:15:31.837
we're also continuing to make rapid

02:15:31.920 --> 02:15:34.717
progress on our long-term goal of

02:15:34.800 --> 02:15:37.797
building a scalable quantum computer.

02:15:37.880 --> 02:15:40.717
Uh we announced uh last year our first

02:15:40.800 --> 02:15:44.917
QPU. Uh we created
a new state of matter uh

02:15:45.000 --> 02:15:47.717
that was only theorized 100 years ago,

02:15:47.800 --> 02:15:50.677
uh and we proved it out that it exists.

02:15:50.760 --> 02:15:53.637
Our vision was to take a very radically

02:15:53.720 --> 02:15:56.917
different approach uh to addressing the

02:15:57.000 --> 02:16:01.317
fundamental barriers uh to building
a scalable quantum machine, which is all

02:16:01.400 --> 02:16:03.237
about reliability,

02:16:03.320 --> 02:16:06.357
uh speed, as well as size. Uh since

02:16:06.440 --> 02:16:09.197
then, we have continued to make progress

02:16:09.280 --> 02:16:11.997
across the full quantum stack uh, with

02:16:12.080 --> 02:16:14.637
both our academic and industry partners.

02:16:14.720 --> 02:16:17.037
In fact, in Q North, we will have a

02:16:17.120 --> 02:16:21.757
quantum computer powered by Adam
computers, natural Adam computers. Uh,

02:16:21.840 --> 02:16:23.837
with our stack in there, we're also

02:16:23.920 --> 02:16:25.997
working with algorithmic and Columbia

02:16:26.080 --> 02:16:31.677
and ETH Zurich. Um, and we
ourselves in fact use this discovery

02:16:31.760 --> 02:16:32.317
um,

02:16:32.400 --> 02:16:35.917
agentic loop to accelerate the work

02:16:36.000 --> 02:16:38.797
uh, in quantum compressing years of

02:16:38.880 --> 02:16:41.797
research uh, into this last year. And

02:16:41.880 --> 02:16:45.717
today, I'm really thrilled to announce

02:16:45.800 --> 02:16:50.037
uh, Myrana 2. And so, this is Myrana 2.

02:16:50.120 --> 02:16:53.437
Um, Myrana 2 implements

02:16:53.520 --> 02:16:56.517
the next-generation material stack that

02:16:56.600 --> 02:16:59.837
we use discovery to discover and build

02:16:59.920 --> 02:17:02.637
and help fabricate. The resulting qubits

02:17:02.720 --> 02:17:05.197
are exceptionally reliable and capable

02:17:05.280 --> 02:17:08.117
of maintaining uh, their state much

02:17:08.200 --> 02:17:11.037
longer. While other common approaches

02:17:11.120 --> 02:17:13.557
deliver a lifetime of just microseconds

02:17:13.640 --> 02:17:16.797
or even milliseconds, Myrana 2 provides

02:17:16.880 --> 02:17:20.917
qubit mean lifetime of 20 seconds or up

02:17:21.000 --> 02:17:23.237
to even a minute. Uh, essentially a

02:17:23.320 --> 02:17:29.087
thousand times higher than what we
were able to achieve with Myrana 1. Um,

02:17:29.170 --> 02:17:32.677
>> [applause] >> and the operations,
and this is so key,

02:17:32.760 --> 02:17:35.397
operations in Myrana 2 are 1

02:17:35.480 --> 02:17:38.157
microsecond, enabling pretty complex

02:17:38.240 --> 02:17:40.877
compact you know, quantum computation

02:17:40.960 --> 02:17:43.397
uh, in that lifetime. And all of this in

02:17:43.480 --> 02:17:47.117
the same qubit form factor of Myrana 1.

02:17:47.200 --> 02:17:51.317
Uh, in 1 100th of a millimeter
controlled all digitally, which is again

02:17:51.400 --> 02:17:54.077
a super important aspect. Uh, making it

02:17:54.160 --> 02:17:56.477
all possible to fit a million of these

02:17:56.560 --> 02:17:59.437
qubits in a chip smaller than a credit

02:17:59.520 --> 02:18:02.317
card. It's this combination

02:18:02.400 --> 02:18:05.357
of the reliability, the speed, the size

02:18:05.440 --> 02:18:08.117
that makes the topological approach so

02:18:08.200 --> 02:18:11.037
unique. Uh with Myrana 1, we had proven

02:18:11.120 --> 02:18:13.797
out the foundational physics, and with

02:18:13.880 --> 02:18:16.957
Myrana 2, now we begin the engineering

02:18:17.040 --> 02:18:22.930
scale. But ultimately, yeah. >> [applause]

02:18:24.360 --> 02:18:28.477
>> But ultimately, it is never about tech

02:18:28.560 --> 02:18:30.277
for tech's sake.

02:18:30.360 --> 02:18:32.837
It's about tackling those pressing

02:18:32.920 --> 02:18:36.757
challenges of people and planet.

02:18:36.840 --> 02:18:39.157
It is also the fundamental point of this

02:18:39.240 --> 02:18:44.117
conference. The question is not whether
you can build the next great model, the

02:18:44.200 --> 02:18:47.877
next great platform,
or even this quantum machine.

02:18:47.960 --> 02:18:50.397
It's the question is how do we build

02:18:50.480 --> 02:18:53.717
this frontier ecosystem together?

02:18:53.800 --> 02:18:56.397
Because there are really two stories

02:18:56.480 --> 02:18:59.357
people can tell about this moment.

02:18:59.440 --> 02:19:01.957
One is that technology concentrates

02:19:02.040 --> 02:19:05.397
power, reduces human agency, and leaves

02:19:05.480 --> 02:19:09.077
the society to absorb the consequences.

02:19:09.160 --> 02:19:12.397
The other is that we use this next wave

02:19:12.480 --> 02:19:16.877
to unlock opportunity for developers,
scientists, enterprises, and every

02:19:16.960 --> 02:19:20.557
community. And our job
is to make the second story

02:19:20.640 --> 02:19:23.437
true. It's That's our North Star for the

02:19:23.520 --> 02:19:28.077
frontier ecosystem. Let's all go build
together. Thank you all very, very much.

02:19:28.160 --> 02:19:31.050
Thank you. >> [applause and cheering]