WEBVTT 00:00:00.080 --> 00:00:03.517 A billion people every [music] single week use chatbt. Like a billion people. 00:00:03.600 --> 00:00:07.357 No one else is doing that in human operated software. The response is too 00:00:07.440 --> 00:00:09.997 slow. I feel like this can be the [clears throat] most empowering [music] 00:00:10.080 --> 00:00:14.040 technology that is ever created. 00:00:14.400 --> 00:00:18.797 >> I think it would be helpful to start with just the news this week. So Denise 00:00:18.880 --> 00:00:23.517 Dresser, your CRO leaving the company, you bringing in a new one, Brad Lap 00:00:23.600 --> 00:00:25.357 leaving. Um, there are a lot of 00:00:25.440 --> 00:00:30.877 questions publicly about OpenAI's leadership bench and what's going on and 00:00:30.960 --> 00:00:33.437 you've come in and are now managing a lot of this and I'd love to hear from 00:00:33.520 --> 00:00:37.917 you. What are all these departures connected? Are they different? What's 00:00:38.000 --> 00:00:39.677 what is there any through line there? 00:00:39.760 --> 00:00:42.637 >> Look, the way that I think about it is that the theme of this year has really 00:00:42.720 --> 00:00:46.717 been focus, right? Has really been >> we have line of sight to such incredible 00:00:46.800 --> 00:00:49.677 technology, but it's not enough just to produce the technology. We need to 00:00:49.760 --> 00:00:52.797 really figure out how do we have this technology benefit people? how do we 00:00:52.880 --> 00:00:55.757 bring it into every enterprise in the world in a way that actually uplifts 00:00:55.840 --> 00:00:58.157 what they're doing, helps them accomplish their goals. So that's really 00:00:58.240 --> 00:01:01.997 been the singular focus that we've been pursuing and that in the purpose in the 00:01:02.080 --> 00:01:04.957 pursuit of that that we've been really trying to think about do we have the 00:01:05.040 --> 00:01:07.197 right structure, do we have the right strategy, do we have all the right 00:01:07.280 --> 00:01:11.037 things there and I think that my view of OpenAI is that we're very resilient 00:01:11.120 --> 00:01:14.317 organization. Um but a lot of what we've been doing is really trying to build the 00:01:14.400 --> 00:01:17.037 leadership team that uh can really help 00:01:17.120 --> 00:01:21.677 us achieve that goal. So I think that my view of everyone who has been here and 00:01:21.760 --> 00:01:25.197 people who who are no longer here is that they've all helped to contribute to 00:01:25.280 --> 00:01:29.197 what OpenAI is today. And looking forward I think that we are scaling 00:01:29.280 --> 00:01:31.197 systematically to every enterprise in 00:01:31.280 --> 00:01:33.677 the world. We're bringing our models to 00:01:33.760 --> 00:01:38.957 consumers everywhere and that a lot of my focus has been making sure that we 00:01:39.040 --> 00:01:40.077 can achieve that. 00:01:40.160 --> 00:01:43.757 >> On the CRO side specifically, um what is 00:01:43.840 --> 00:01:47.997 it that you weren't getting that you needed? you know, is is is that the way 00:01:48.080 --> 00:01:50.877 to think about it? >> Look, I would say it differently. Like a 00:01:50.960 --> 00:01:54.477 year ago, I think that we really were not in the game at all, right, when it 00:01:54.560 --> 00:01:57.517 came to enterprise. It just wasn't really our focus. We hadn't really built 00:01:57.600 --> 00:02:01.197 the muscle we needed. And now, I think we actually are in the game. Like, we 00:02:01.280 --> 00:02:04.557 kind of have really built a foundation that I think is actually something we 00:02:04.640 --> 00:02:07.757 can really build on. And now we're systematically scaling. And that that to 00:02:07.840 --> 00:02:10.797 me is like the moment. It's not really about the individuals in some ways. It's 00:02:10.880 --> 00:02:14.157 really about the overall sort of direction of the field. How does that 00:02:14.240 --> 00:02:17.277 change this systematically scaling you're talking about? What does that 00:02:17.360 --> 00:02:19.037 mean practically for open AI? 00:02:19.120 --> 00:02:22.397 >> Well, number one uh is that I think that 00:02:22.480 --> 00:02:26.077 we are looking at like transforming the 00:02:26.160 --> 00:02:29.517 entire economy. Like that's the thing that I think is very hard to see from 00:02:29.600 --> 00:02:32.717 the outside, right? Is it kind of feels like hey, it's a business like we're bu 00:02:32.800 --> 00:02:35.117 selling SAS software. That's not what we're doing, right? We're doing 00:02:35.200 --> 00:02:39.037 something very different. We are transforming the economy to be a compute 00:02:39.120 --> 00:02:42.637 powered one. And I think that what an enterprise even is will change. And so 00:02:42.720 --> 00:02:45.357 the question is well where do you start? 00:02:45.440 --> 00:02:49.677 How do you make sure that you understand every segment and you're able to sell in 00:02:49.760 --> 00:02:51.757 a repeatable fashion to all of them? And 00:02:51.840 --> 00:02:56.717 so a lot of the motion that we've been working on and again this isn't a it's 00:02:56.800 --> 00:02:59.757 it's not it's not a new motion in some ways. is just one that we need to now 00:02:59.840 --> 00:03:03.357 scale just given where the business is at is to really make sure that everyone 00:03:03.440 --> 00:03:06.637 in the field we're able to funnel back the insights that we're seeing from 00:03:06.720 --> 00:03:10.877 customers in a very systematic way back to our product teams our research teams 00:03:10.960 --> 00:03:14.237 and vice versa like one thing that we have learned like if you look at for 00:03:14.320 --> 00:03:17.997 example how we produce blog posts internally that we learned years ago 00:03:18.080 --> 00:03:21.677 that it has to be a collaboration right you cannot have a handoff of oh here's 00:03:21.760 --> 00:03:25.277 the comms people who write blog posts and here's the research people who 00:03:25.360 --> 00:03:28.637 produce models and you know that they'll just like kind of hand hand it off in 00:03:28.720 --> 00:03:31.277 some way. It has to be deep collaboration. You need to have the 00:03:31.360 --> 00:03:33.597 technical expert really hold the pen and 00:03:33.680 --> 00:03:38.317 feel accountability for the end result being good. And they can be amplified by 00:03:38.400 --> 00:03:41.677 people who are experts on how do you actually craft a narrative? How do you 00:03:41.760 --> 00:03:43.037 sort of bring together all these assets? 00:03:43.120 --> 00:03:44.717 How do you have a good launch? 00:03:44.800 --> 00:03:49.037 >> And so that to me is the core activity that we are pursuing is to say we can't 00:03:49.120 --> 00:03:52.157 be fragmented. We can't be working on all the things. We have to be unified 00:03:52.240 --> 00:03:56.477 and focused. And so that's the internal answer for how we're working. And in 00:03:56.560 --> 00:04:00.797 terms of the external, I think that you should see that across that we will sort 00:04:00.880 --> 00:04:03.917 of systematically be prioritizing different industries, really focused on 00:04:04.000 --> 00:04:08.877 the uh the deep understanding of the metrics and exactly where the token 00:04:08.960 --> 00:04:13.037 spend is growing the most and really focused on outcomes. Like every business 00:04:13.120 --> 00:04:15.277 right now is asking where's the ROI. 00:04:15.360 --> 00:04:17.037 Yeah. >> And that is something that is a big 00:04:17.120 --> 00:04:20.477 focus of ours. Ours has always been the hey we have the best price performant 00:04:20.560 --> 00:04:24.557 models, but we really have to show people that in the outcomes for them. um 00:04:24.640 --> 00:04:27.917 also really scaling further through channels and through through partners. 00:04:28.000 --> 00:04:29.197 That's something we believe in as well. 00:04:29.280 --> 00:04:33.357 And so I think that a lot of where we're at is we feel very confident about the 00:04:33.440 --> 00:04:37.277 fundamentals of the motion and the fundamentals of we have a great product, 00:04:37.360 --> 00:04:41.597 a great model and then really just rapidly rapidly growing both our 00:04:41.680 --> 00:04:45.117 internal team so that we can reach everyone but doing that in a way that 00:04:45.200 --> 00:04:48.397 that really has the that positive effect that we're looking for. You use the word 00:04:48.480 --> 00:04:51.677 accountability and I'm curious from your perspective, what are you accountable 00:04:51.760 --> 00:04:54.797 for now at OpenAI because of the leadership changes that have happened? 00:04:54.880 --> 00:04:57.117 Fiji Simo came in for a time. She had to 00:04:57.200 --> 00:04:59.437 step back permanently due to her health. 00:04:59.520 --> 00:05:03.357 You stepped in while she was uh out and took over a large part of of the 00:05:03.440 --> 00:05:06.397 company. Now it's you and Sam effectively running it together. How was 00:05:06.480 --> 00:05:09.837 that divided up? >> So for the past couple years, I've been 00:05:09.920 --> 00:05:14.317 accountable for the infrastructure side of what we're doing. So data centered, 00:05:14.400 --> 00:05:18.477 compute, those kinds of things. And that that part remains um that there's a 00:05:18.560 --> 00:05:22.717 large part of the scope that I was also working on in terms of the software for 00:05:22.800 --> 00:05:25.837 kernels and sort of how you train these big models that that's something that 00:05:25.920 --> 00:05:29.437 I've I've been able to hand off so that I can really focus on the business side. 00:05:29.520 --> 00:05:30.957 And so I would view me as sort of 00:05:31.040 --> 00:05:33.757 endto-end accountable for revenue for 00:05:33.840 --> 00:05:37.597 product marketing the whole machine for 00:05:37.680 --> 00:05:41.357 bringing these models from research to 00:05:41.440 --> 00:05:45.357 >> value for our customers. When you look at this machine, and I know you're still 00:05:45.440 --> 00:05:48.477 putting pieces in place, but when you look at this machine you have in front 00:05:48.560 --> 00:05:52.237 of you that you're accountable for, how is it doing relative to competition? I 00:05:52.320 --> 00:05:55.837 mean, I think specifically Anthropic, there's reporting that their ARR is I 00:05:55.920 --> 00:05:59.997 mean, it's potentially more than double what Open AIS is. Um, what do you what 00:06:00.080 --> 00:06:04.557 do you make of that? >> Well, I think we started late and 00:06:04.640 --> 00:06:08.797 specifically the way to think about the overall field is that for the past 00:06:08.880 --> 00:06:12.397 couple years, the chat market was the main growing one. And the reason for 00:06:12.480 --> 00:06:16.157 this was simple. GPD4 was the first model where the results were 00:06:16.240 --> 00:06:19.997 personalized enough that it made sense for you to ask a question, read the 00:06:20.080 --> 00:06:21.757 answer, it's interesting to you. And so 00:06:21.840 --> 00:06:27.197 it very naturally lends itself to a chat, right, to a chat bot. And that is 00:06:27.280 --> 00:06:31.037 something we're able to scale, deliver value. Now a billion people every single 00:06:31.120 --> 00:06:34.957 week use chatbt. Like a billion people like that is scale. That is real scale. 00:06:35.040 --> 00:06:38.877 No one else is doing that. But this year something else happened is the agentic 00:06:38.960 --> 00:06:41.037 moment. This really was December of of 00:06:41.120 --> 00:06:43.677 last year is when both we and 00:06:43.760 --> 00:06:48.557 competition had a moment where agents were really able to first start doing 00:06:48.640 --> 00:06:51.917 software engineering in a significant way. It went from saying that you know 00:06:52.000 --> 00:06:55.437 you could speed up your work with maybe 20% 00:06:55.520 --> 00:06:59.517 productive productivity increased to 80% of the work could now be done with an 00:06:59.600 --> 00:07:04.637 AI. And so suddenly the way that software engineers operated it inverted 00:07:04.720 --> 00:07:09.117 that you became much more of a manager of agents. And since then the growth has 00:07:09.200 --> 00:07:11.837 been unprecedented absolutely unprecedented how these agents have been 00:07:11.920 --> 00:07:12.797 adopted. 00:07:12.880 --> 00:07:14.717 >> And one thing that happened was I think 00:07:14.800 --> 00:07:21.037 that we focused late on coding in particular actually let me let me re 00:07:21.120 --> 00:07:24.317 okay >> let me let me let me sort of give a different answer as well. So there's 00:07:24.400 --> 00:07:29.197 there's one simple narrative which is that we focused less on real world 00:07:29.280 --> 00:07:32.717 coding >> and we came to that game late because we 00:07:32.800 --> 00:07:36.557 always had looked at things like the coding competition benchmarks and always 00:07:36.640 --> 00:07:40.797 had the lead there but less the okay but how is a developer really going to use 00:07:40.880 --> 00:07:45.117 it like the messy real world codebase the developer interrupting in the middle 00:07:45.200 --> 00:07:49.037 what's the personality of the model these sort of little last mile paper 00:07:49.120 --> 00:07:53.037 cuts that actually make a huge difference in adoption and similar on 00:07:53.120 --> 00:07:56.717 the GTM side of really thinking about how do you show how do you build 00:07:56.800 --> 00:08:00.317 relationships with customers? How do you build the kind of observability controls 00:08:00.400 --> 00:08:03.517 that they need? How do you help them understand what's possible within their 00:08:03.600 --> 00:08:04.557 context? 00:08:04.640 --> 00:08:07.757 >> And all of these were less our singular 00:08:07.840 --> 00:08:10.557 focus until much later than competition. 00:08:10.640 --> 00:08:14.237 So I think that we started late but our trajectory has been very steep. But 00:08:14.320 --> 00:08:18.477 there's actually even a deeper answer that is less obvious from the outside 00:08:18.560 --> 00:08:21.037 and that is one of model capability. For 00:08:21.120 --> 00:08:23.997 a period of about two years, right from 00:08:24.080 --> 00:08:27.917 when we announced GBD40 in April of 2024 00:08:28.000 --> 00:08:31.277 to December of 2025, we were really 00:08:31.360 --> 00:08:34.237 squeezing the juice out of the GPD40 00:08:34.320 --> 00:08:36.717 model, right? that that we sort of had a 00:08:36.800 --> 00:08:39.437 bunch of pieces that uh we were 00:08:39.520 --> 00:08:41.757 improving but that fundamentally we were 00:08:41.840 --> 00:08:46.717 on the same underlying base model the same underlying pre-train and there's a 00:08:46.800 --> 00:08:50.637 lot of reasons right that is really because to produce and ship a pre-train 00:08:50.720 --> 00:08:53.917 is something that requires full integration across the company it's not 00:08:54.000 --> 00:08:57.277 any one team it's about the integration across these teams to think about the 00:08:57.360 --> 00:08:59.517 inference to think about the post training to make sure you've co-designed 00:08:59.600 --> 00:09:03.197 the model to be the right size all of these things you have to get exactly 00:09:03.280 --> 00:09:06.157 right and have a a sort of harmonious way of working. 00:09:06.240 --> 00:09:08.797 >> And over the course of 2025, a lot of 00:09:08.880 --> 00:09:10.637 our focus, a lot of my personal focus 00:09:10.720 --> 00:09:13.277 was how do we really get back to 00:09:13.360 --> 00:09:17.677 shipping new pre-trains as a core part of our innovation engine. 00:09:17.760 --> 00:09:22.477 >> And this year, you see the results. So, we've actually shipped already. We're on 00:09:22.560 --> 00:09:26.077 our third pre-train. We're actually about to ship that. Um, >> this is Astra. 00:09:26.160 --> 00:09:30.157 >> This is Astra. >> Yeah. >> And a lot of the gains that you see are 00:09:30.240 --> 00:09:33.197 because we have made the model so fundamentally better. And so the way to 00:09:33.280 --> 00:09:36.797 think about it is that there's an underlying technological place where we 00:09:36.880 --> 00:09:40.957 really were lagging. There was a kind of a core part of our innovation machine 00:09:41.040 --> 00:09:45.757 that we weren't able to fully bring to bear. But now it's different. 00:09:45.840 --> 00:09:49.597 >> And that is why you see our model capabilities taking off so much. And if 00:09:49.680 --> 00:09:53.277 you look at 56 soul when we launched that best model in the market, you could 00:09:53.360 --> 00:09:57.917 see the the revenue really starting to shift. We grew 20% in our topline 00:09:58.000 --> 00:10:01.997 revenue in July. >> Mh. >> And we see that exponential continuing. 00:10:02.080 --> 00:10:05.757 M another thing that's changed as you were saying with the model capability is 00:10:05.840 --> 00:10:08.717 the models are becoming incredibly 00:10:08.800 --> 00:10:13.197 capable at cyber and uh at hacking and I 00:10:13.280 --> 00:10:17.517 think I mean the world is still kind of thinking through the implications of the 00:10:17.600 --> 00:10:20.957 hugging fa hugging face incident. I mean what essentially happened I mean you 00:10:21.040 --> 00:10:25.117 correct me if I'm wrong but it seems what happened is you all accidentally 00:10:25.200 --> 00:10:26.957 hacked another company without knowing 00:10:27.040 --> 00:10:29.517 it for a while and it was a model that 00:10:29.600 --> 00:10:34.397 you all were training and evaluating and you've since you know there was a black 00:10:34.480 --> 00:10:37.917 hat video um where you guys walked through this and you're going to put 00:10:38.000 --> 00:10:42.557 more out in the coming days um I think but walk me through that when you heard 00:10:42.640 --> 00:10:46.077 that that had happened what did you think 00:10:46.160 --> 00:10:48.797 >> well in the words of someone on our 00:10:48.880 --> 00:10:52.077 security team. It was definitely the 00:10:52.160 --> 00:10:54.717 sort of most eye-opening 00:10:54.800 --> 00:10:57.837 security event that any of us were aware 00:10:57.920 --> 00:11:01.917 of. The way to think about it is that there's a lot of different lessons that 00:11:02.000 --> 00:11:06.317 we're taking away from it. But I think that it is a moment where we can as a 00:11:06.400 --> 00:11:10.397 community learn. We can understand what the shape of the future is. And to me 00:11:10.480 --> 00:11:12.237 that the salient points for the purpose 00:11:12.320 --> 00:11:14.797 of how how I'm thinking about it right 00:11:14.880 --> 00:11:19.917 now is that we had a model and we were evaluating it and it managed to it was 00:11:20.000 --> 00:11:24.397 in a sandbox that was well configured and it managed to both break out of the 00:11:24.480 --> 00:11:28.397 sandbox and then also break into production infrastructure. it was 00:11:28.480 --> 00:11:32.397 intended to be >> hardened. >> And so that the capability there is 00:11:32.480 --> 00:11:37.037 quite high and that there's some details about exactly how it all worked and the 00:11:37.120 --> 00:11:39.997 fact that it was kind of not just a singular agent but really a swarm of 00:11:40.080 --> 00:11:44.317 agents that were focused on on this goal and able to pass messages to each other. 00:11:44.400 --> 00:11:46.957 And the sort of thing that you start to 00:11:47.040 --> 00:11:50.317 see is that with very capable agents 00:11:50.400 --> 00:11:54.877 that there are limitations that human attackers have that slow us down that 00:11:54.960 --> 00:11:58.317 will no longer hold back the attackers, right? that they can look through lots 00:11:58.400 --> 00:12:01.997 of information through many different little vulnerabilities and chain them 00:12:02.080 --> 00:12:04.557 together into something significant. For 00:12:04.640 --> 00:12:09.357 me, what I immediately started to realize and think about is what are the 00:12:09.440 --> 00:12:13.437 implications for six months from now when this kind of capability isn't just 00:12:13.520 --> 00:12:18.877 in the lab, isn't just in a few places, but is really widely available. And I 00:12:18.960 --> 00:12:22.797 think that the answer is that every company needs to respond, needs to 00:12:22.880 --> 00:12:25.197 upgrade its security because attackers 00:12:25.280 --> 00:12:27.437 are going to fully automate and you can 00:12:27.520 --> 00:12:32.797 even see with the release of GLM 5.3 that the cyber capabilities of 00:12:32.880 --> 00:12:36.717 openweight models available to everyone are going to increase and I think that 00:12:36.800 --> 00:12:40.637 the important thing is that I actually think in the long run that defenders can 00:12:40.720 --> 00:12:44.397 win that AI can actually make it so defenders can decisively win in a way 00:12:44.480 --> 00:12:47.677 that was never possible before right because security always a cat-and- mouse 00:12:47.760 --> 00:12:50.557 Attackers up level, defenders up level. 00:12:50.640 --> 00:12:55.117 But AI can give defenders an unfair advantage. For example, formal 00:12:55.200 --> 00:12:58.077 verification of software where you can actually prove that is secure, 00:12:58.160 --> 00:13:00.477 mathematically prove security properties. 00:13:00.560 --> 00:13:04.557 >> The technology for this exists, but it is interactable for humans to actually 00:13:04.640 --> 00:13:06.877 do the work to prove those properties. 00:13:06.960 --> 00:13:09.357 >> And we have these models that are incredible at math, right? They are 00:13:09.440 --> 00:13:13.757 solving, for example, Astra, our upcoming model solved, we we announced 00:13:13.840 --> 00:13:18.397 this a few weeks ago, solved 10 math problems. And this was with the $2,000 00:13:18.480 --> 00:13:21.517 worth of compute that had been open for decades, right? Mathematicians that 00:13:21.600 --> 00:13:24.157 spent their whole careers on these problems and now we have a model that 00:13:24.240 --> 00:13:26.317 can prove the answers. Those models 00:13:26.400 --> 00:13:31.277 definitely can prove formal safety properties of software. 00:13:31.360 --> 00:13:35.357 >> And so that's one place where unfair advantage for for defenders. A second 00:13:35.440 --> 00:13:39.757 thing that we're doing is we're really investing in generating secure code. 00:13:39.840 --> 00:13:41.277 >> And you start to realize that as a human 00:13:41.360 --> 00:13:45.277 developer, there are many patterns that people just keep getting wrong. And I've 00:13:45.360 --> 00:13:49.357 been through this many times in my previous jobs. I was very focused on 00:13:49.440 --> 00:13:53.277 building secure infrastructure and just find like all these little things people 00:13:53.360 --> 00:13:56.717 do wrong. SQL injections people kind of know about that there's little patterns, 00:13:56.800 --> 00:14:00.877 but there's lots of little anti-atterns as well for even like for example 00:14:00.960 --> 00:14:04.397 >> I remember it used to be like in Ruby that the regular expressions there's 00:14:04.480 --> 00:14:08.317 like certain characters that if you get them wrong that that there's just very 00:14:08.400 --> 00:14:11.437 common attacks uh where people can put a new line and it doesn't match. So you're 00:14:11.520 --> 00:14:13.757 able to to to add things to the payload. 00:14:13.840 --> 00:14:17.677 all of those problems of just little security antiatterns that you just keep 00:14:17.760 --> 00:14:21.037 having to tell your people don't do this, don't do this, don't do this, you 00:14:21.120 --> 00:14:24.077 can train your AI to never do it. And so I think that we actually have the 00:14:24.160 --> 00:14:25.517 ability to build the best security 00:14:25.600 --> 00:14:29.037 engineers in the world and fundamentally 00:14:29.120 --> 00:14:32.957 the defenders can have the advantage, 00:14:33.040 --> 00:14:37.357 but this is going to require a mindset shift. This is going to require a mega 00:14:37.440 --> 00:14:41.357 project. This is something that the whole ecosystem needs to change how 00:14:41.440 --> 00:14:44.797 software is secured because there's a second thing that we've learned which is 00:14:44.880 --> 00:14:46.877 that as these cyber models have gotten 00:14:46.960 --> 00:14:49.517 more capable the number of bugs that 00:14:49.600 --> 00:14:53.117 they are finding in critical software 00:14:53.200 --> 00:14:57.757 for example the Linux kernel they just keep finding more and more and it really 00:14:57.840 --> 00:15:01.517 makes you realize that human written software yeah that the response is too 00:15:01.600 --> 00:15:03.997 slow. Yeah. And so we need defenders to 00:15:04.080 --> 00:15:09.197 automate, to move at machine speed because their adversaries certainly 00:15:09.280 --> 00:15:11.917 will. >> What do you say to people who maybe and 00:15:12.000 --> 00:15:15.197 also they already have a negative kind of disposition of AI, maybe there's a 00:15:15.280 --> 00:15:18.397 data center they don't like in their neighborhood or something and they hear 00:15:18.480 --> 00:15:23.677 you say that and go, you're just selling the answer to the problem you've created 00:15:23.760 --> 00:15:26.077 in the sense of like you're building these models and now you're going to 00:15:26.160 --> 00:15:30.077 sell something to like protect me from them. What would you say to that? Well, 00:15:30.160 --> 00:15:31.917 f first of all, I have a lot of empathy 00:15:32.000 --> 00:15:37.517 for that reaction. The way that I look at it is I rewind to 10 years ago, 11 00:15:37.600 --> 00:15:41.437 years ago, before we started this company, I wasn't in AI. The reason I 00:15:41.520 --> 00:15:45.037 wanted to be an AI is because I feel like this can be the most empowering 00:15:45.120 --> 00:15:48.397 technology that has ever created. That AI can be something that helps everyone 00:15:48.480 --> 00:15:51.437 live the life that they want, build the life that they want, achieve their 00:15:51.520 --> 00:15:55.277 goals. And I just wanted to help steer it in a slightly more positive direction 00:15:55.360 --> 00:15:58.317 than it would be without me. So I started because I feel like this 00:15:58.400 --> 00:16:00.717 technology can uplift everyone and when 00:16:00.800 --> 00:16:02.797 you look at where open AI is and what 00:16:02.880 --> 00:16:06.317 we're doing we can see a little bit into 00:16:06.400 --> 00:16:10.477 the future right that what we are creating it's not just us right that 00:16:10.560 --> 00:16:13.677 there's a whole field there's a whole ecosystem there are many actors and that 00:16:13.760 --> 00:16:17.117 is itself a good thing that we believe that there should be democratization of 00:16:17.200 --> 00:16:21.037 this technology it should not be locked up in one place with one entity 00:16:21.120 --> 00:16:24.797 controlling it because that's a different kind of risk but it means that 00:16:24.880 --> 00:16:27.757 we do need to adapt we do need navigate through and we need to use the 00:16:27.840 --> 00:16:31.997 information that we have sort of painstakingly extracted from the future 00:16:32.080 --> 00:16:34.157 to see what is the shape of it. So to me 00:16:34.240 --> 00:16:38.797 hugging face is a sort of proof point a 00:16:38.880 --> 00:16:42.077 case study a window into the future. 00:16:42.160 --> 00:16:45.357 It's not necessarily a future that any one company is creating. It's the 00:16:45.440 --> 00:16:48.877 overall economy the overall set of actors in all these different countries 00:16:48.960 --> 00:16:52.077 who are building these kinds of models and it shows you what will be possible 00:16:52.160 --> 00:16:56.157 and how attackers will be able to update. What it doesn't show you is how 00:16:56.240 --> 00:16:58.797 do defenders update? But that to me is 00:16:58.880 --> 00:17:03.597 what we are also trying to now show as well that we can build these models that 00:17:03.680 --> 00:17:06.957 you can use to scan yourself. As an example, 00:17:07.040 --> 00:17:11.837 >> I asked codecs and this is just off-the-shelf codecs, no special access, 00:17:11.920 --> 00:17:15.117 no special models. I just use my personal codecs to say 00:17:15.200 --> 00:17:18.477 >> scan gregarman.com. Tell me what vulnerabilities are there. 00:17:18.560 --> 00:17:22.397 >> And gregarman.com is very simple. It's just a static site. I was like, come on. 00:17:22.480 --> 00:17:26.637 What what's it going to find? And it came back with like 13 findings. And 00:17:26.720 --> 00:17:29.997 each one of them I was like, you know, I kind of knew about this, but it's so 00:17:30.080 --> 00:17:33.917 much work to fix it. And like, oh yeah, I kind of forgot that I left that open. 00:17:34.000 --> 00:17:37.277 Or, you know, here's some domain that was still, you know, was was legacy. I'm 00:17:37.360 --> 00:17:40.397 not using it anymore. I hadn't set the records that prevent people from 00:17:40.480 --> 00:17:43.117 spoofing your email address. A bunch of little things like that. 00:17:43.200 --> 00:17:44.477 >> That one seems like a big deal. 00:17:44.560 --> 00:17:49.917 >> Well, yeah, you're right. Now that you say it. Now, here's the thing. Honestly, 00:17:50.000 --> 00:17:52.717 to fix those kinds of things, you have to set up like Demark to do it right. 00:17:52.800 --> 00:17:55.517 And there's this like whole like process you have to do to roll. I'm just like I 00:17:55.600 --> 00:17:58.077 don't know it just takes time. It takes a lot of expertise. So instead I just 00:17:58.160 --> 00:17:59.597 asked Codex, can you fix it really? 00:17:59.680 --> 00:18:03.277 >> And so it opened up my cloud for a control panel and it just used computer 00:18:03.360 --> 00:18:05.837 use it. It clicked around it like set a bunch of headers changed a bunch of 00:18:05.920 --> 00:18:09.037 settings. I used to have it so that I had Cloudflare in front and then AWS on 00:18:09.120 --> 00:18:13.197 the back end. I was going over insecure HTTP here. Um Codex had figured that 00:18:13.280 --> 00:18:16.557 out. I actually not even sure how it how it saw that because I thought everything 00:18:16.640 --> 00:18:19.357 was hidden behind Cloudflare. It actually was just like you know what I'm 00:18:19.440 --> 00:18:23.837 just going to migrate you off of AWS. So it set up Cloudflare pages, moved over 00:18:23.920 --> 00:18:27.837 the content. I had an old jQuery version that was insecure. It just got rid of 00:18:27.920 --> 00:18:32.477 jQuery entirely. And it was like 15 minutes to find all these findings. It 00:18:32.560 --> 00:18:36.237 was another hour to fix all of them. And for me, I just was the manager. I just 00:18:36.320 --> 00:18:39.037 gave it the intent. I said, "What I want is I want to know where the problems are 00:18:39.120 --> 00:18:43.037 and I want you to fix them." And it did it. And that to me is more the future 00:18:43.120 --> 00:18:45.757 that we're promised. >> I want to hear, and I know this is more 00:18:45.840 --> 00:18:49.517 in the research world, but I think it it's really relevant to everything. Um 00:18:49.600 --> 00:18:51.277 you talk about what you guys are doing 00:18:51.360 --> 00:18:55.677 now in response to hugging face. Um one 00:18:55.760 --> 00:18:59.837 of your colleagues Mia told me the other day that um I wish we had done a lot of 00:18:59.920 --> 00:19:01.437 the work we're doing before this 00:19:01.520 --> 00:19:03.677 happened. Uh so what is that work that's 00:19:03.760 --> 00:19:07.117 happening now? So I would view the mode 00:19:07.200 --> 00:19:09.277 we've been in since the discovery of the 00:19:09.360 --> 00:19:11.677 hugging face incident as full incident 00:19:11.760 --> 00:19:15.437 response to uplevel every aspect of how 00:19:15.520 --> 00:19:20.637 we secure, monitor and align our most capable models. 00:19:20.720 --> 00:19:22.557 >> Yeah. >> And we believe that models that are 00:19:22.640 --> 00:19:24.397 above a certain capability level, they 00:19:24.480 --> 00:19:25.677 require this. 00:19:25.760 --> 00:19:29.837 >> And the it's been a it's been a very 00:19:29.920 --> 00:19:31.837 interesting technical problem to think 00:19:31.920 --> 00:19:35.357 about. And it's not all us being caught 00:19:35.440 --> 00:19:38.877 flatfooted because I mean we anticipated 00:19:38.960 --> 00:19:42.797 this moment for quite some time. If you look back at February publicly we 00:19:42.880 --> 00:19:46.877 started talking about trusted access for cyber models, right? That that was the 00:19:46.960 --> 00:19:49.357 first time we were really thinking about okay we've always had in our 00:19:49.440 --> 00:19:52.317 preparedness framework when cyber capabilities get to a certain level you 00:19:52.400 --> 00:19:55.277 need to treat them differently. Now we're starting to think about the fact 00:19:55.360 --> 00:19:57.837 that these are going to be so transformative in the world. We need to 00:19:57.920 --> 00:20:01.917 think about how access is controlled internally. We started to think about if 00:20:02.000 --> 00:20:06.717 we have very capable models that are able to break out of sandboxes, well, we 00:20:06.800 --> 00:20:10.717 need to make sure our sandbox is as industrial-grade as, you know, sort of 00:20:10.800 --> 00:20:12.797 foolproof as humanly possible, which 00:20:12.880 --> 00:20:14.957 really means engineering from the ground 00:20:15.040 --> 00:20:17.917 up around security properties. And so we 00:20:18.000 --> 00:20:22.077 put some of our our um actually Mike Dalton who uh was one of the people who 00:20:22.160 --> 00:20:25.997 gave that Black Hat talk um the the task that he started to work on was build the 00:20:26.080 --> 00:20:29.437 most secure possible sandbox. So, we've been putting in place a bunch of 00:20:29.520 --> 00:20:33.837 infrastructure primitives and policy primitives for this moment, but it's 00:20:33.920 --> 00:20:36.477 also different when it actually hits, right? You never know when you're going 00:20:36.560 --> 00:20:40.397 to hit that that moment. And I think here we're actually very I think it's we 00:20:40.480 --> 00:20:42.317 got to see this window in a very 00:20:42.400 --> 00:20:44.557 concrete way of okay, we're now at a 00:20:44.640 --> 00:20:48.397 serious capability level relative to 00:20:48.480 --> 00:20:50.557 the pieces. We must act now. Mhm. 00:20:50.640 --> 00:20:54.957 >> And so there's been a workflow shift that's been happening very rapidly 00:20:55.040 --> 00:20:58.317 because you start to think about >> a lot of our governance has been focused 00:20:58.400 --> 00:21:01.357 on deployment of model but not necessarily the training of model. Now 00:21:01.440 --> 00:21:04.477 we need to pull the governance back in during the training and evaluation 00:21:04.560 --> 00:21:07.997 process. We also need good governance there. We need to have standards around 00:21:08.080 --> 00:21:11.837 checks and balances around how do we know if it's secure enough? How do we 00:21:11.920 --> 00:21:14.797 know if it's aligned enough? And this 00:21:14.880 --> 00:21:16.237 alignment piece is actually very 00:21:16.320 --> 00:21:19.917 important because our belief is that if 00:21:20.000 --> 00:21:22.797 you make models extremely capable, you 00:21:22.880 --> 00:21:28.077 really want to make sure that as the capability grows that their alignment 00:21:28.160 --> 00:21:30.637 grows with them, >> right? That you want those things to be 00:21:30.720 --> 00:21:34.237 in lock step and sort of coordinated fashion. And so in that way you kind of 00:21:34.320 --> 00:21:37.997 have this defense and depth of models that are trying to do the right thing 00:21:38.080 --> 00:21:41.517 that you have sandbox that's as secure as possible and that I think that doing 00:21:41.600 --> 00:21:45.357 this in an iterative fashion so that you're taking incremental jumps so you 00:21:45.440 --> 00:21:49.357 can kind of understand as you go and make sure that your systems are are 00:21:49.440 --> 00:21:52.877 working all of that is very aligned with how we've thought about iterative 00:21:52.960 --> 00:21:55.757 deployment how we've thought about the creation of this technology from the 00:21:55.840 --> 00:21:59.517 very beginning. So we are operationalizing that and actually for 00:21:59.600 --> 00:22:04.077 me as a as a manager as a sort of thinking about the organizational system 00:22:04.160 --> 00:22:08.717 for producing this technology. It's also been a very important challenge because 00:22:08.800 --> 00:22:11.517 this means you now need different people in the room to make decisions. You need 00:22:11.600 --> 00:22:14.797 a different process for how do you know what the requirements are? How do you 00:22:14.880 --> 00:22:18.317 execute against those priorities? If there's a change in your technical 00:22:18.400 --> 00:22:20.877 understanding, how do you propagate that? How do you make sure people in the 00:22:20.960 --> 00:22:24.877 right rooms? But again, in some ways, it's been it's been a phase shift, but 00:22:24.960 --> 00:22:26.317 it's been one we've been preparing for. 00:22:26.400 --> 00:22:31.117 Like earlier this year, I spent a lot of effort to really try to bring together 00:22:31.200 --> 00:22:34.317 research and security to >> start having we we started doing, for 00:22:34.400 --> 00:22:38.077 example, these weekly uh safety and security of agents meetings. And some of 00:22:38.160 --> 00:22:40.477 these things are about build the connective tissue, have people who 00:22:40.560 --> 00:22:42.957 wouldn't normally talk to each other because their worlds are a little bit 00:22:43.040 --> 00:22:44.877 different. They come from different backgrounds, they're focused on 00:22:44.960 --> 00:22:48.157 different problems, be forced to interact, be forced to think about 00:22:48.240 --> 00:22:50.877 shared problems. And a lot of that shared empathy, that shared 00:22:50.960 --> 00:22:55.837 understanding, that shared technical context is paying dividends right now. 00:22:55.920 --> 00:22:59.597 >> At the same time though, the implications of this are that uh 00:22:59.680 --> 00:23:04.237 momentum on new models is going to what significantly slow. Do you have a sense 00:23:04.320 --> 00:23:07.437 of the impact? >> Well, I would put this differently 00:23:07.520 --> 00:23:11.837 because I've always taken this lens to how do you think about progress? How do 00:23:11.920 --> 00:23:13.357 you think about progress in this field? 00:23:13.440 --> 00:23:15.997 And if you only think about progress as 00:23:16.080 --> 00:23:18.397 what's your score on a benchmark, what's 00:23:18.480 --> 00:23:20.397 your overall capability level in this 00:23:20.480 --> 00:23:24.077 specific task, that's too myopic. We are 00:23:24.160 --> 00:23:29.117 building an integrated system and safety and security are a core property of that. 00:23:29.200 --> 00:23:32.557 >> And so if you have safety and security lagging relative to the capability, 00:23:32.640 --> 00:23:33.757 yeah, you absolutely need to improve 00:23:33.840 --> 00:23:35.517 your safety and security. But our goal 00:23:35.600 --> 00:23:39.837 is to have a system that we are building 00:23:39.920 --> 00:23:42.797 at full force on safety and security as 00:23:42.880 --> 00:23:44.957 first class requirements. And so the way 00:23:45.040 --> 00:23:47.117 I would think about this is that we are 00:23:47.200 --> 00:23:49.997 upleveling every aspect of the machine. 00:23:50.080 --> 00:23:54.237 We have spent the past year, past two years upleveling how we build our 00:23:54.320 --> 00:23:57.997 clusters, right? We now are able to train at massive, massive scale. We're 00:23:58.080 --> 00:24:00.717 able to get the whole pre-training 00:24:00.800 --> 00:24:05.597 machine, the RL machine, all of these pieces to work together in concert. We 00:24:05.680 --> 00:24:09.357 are building the best sandbox. We are adding chain of thought monitoring to 00:24:09.440 --> 00:24:14.157 all of this. To me, it's just saying what is the scope of delivery? What does 00:24:14.240 --> 00:24:17.037 success look like? What does good look like? And so the way I would think about 00:24:17.120 --> 00:24:19.997 it is that yeah, if you jettison some requirements, you can move faster in 00:24:20.080 --> 00:24:22.557 other requirements. But that doesn't make any sense. That never makes any 00:24:22.640 --> 00:24:25.517 sense. That's not how engineering works. 00:24:25.600 --> 00:24:29.277 And so to me it's we are pursuing the 00:24:29.360 --> 00:24:33.357 plan like we've really thought about this many times over the past and you 00:24:33.440 --> 00:24:36.397 can see it in many of our writings from the charter to the preparedness 00:24:36.480 --> 00:24:38.077 framework and many other things in 00:24:38.160 --> 00:24:41.437 between that as models get more capable 00:24:41.520 --> 00:24:46.237 you have to make sure that you increase your safety and security requirements in 00:24:46.320 --> 00:24:49.917 concert and that's exactly what we're doing. But I mean practically yes I hear 00:24:50.000 --> 00:24:54.317 all that but it does seem like what this will result in is the release cadence of 00:24:54.400 --> 00:24:58.477 models to lengthen. >> I I'm not certain if that will be true 00:24:58.560 --> 00:25:03.197 to be honest. So I think that what will happen is it does mean that if you were 00:25:03.280 --> 00:25:05.277 just maxing a particular metric you 00:25:05.360 --> 00:25:07.597 could probably max that metric faster 00:25:07.680 --> 00:25:10.717 than what we will pursue. But there's a 00:25:10.800 --> 00:25:13.037 lot of degrees of freedom of hey we have 00:25:13.120 --> 00:25:15.277 we always have a bunch of of um wins 00:25:15.360 --> 00:25:18.877 queued up and that we can jettison some 00:25:18.960 --> 00:25:21.037 of those wins right on particular axis. 00:25:21.120 --> 00:25:23.677 So if you're pushing on particular, you 00:25:23.760 --> 00:25:26.637 know, capabilities on cyber or on bio or 00:25:26.720 --> 00:25:31.037 on coding or any of these things, maybe you don't couple together all of the 00:25:31.120 --> 00:25:34.157 particular things. Or if there's architectural improvements, maybe you 00:25:34.240 --> 00:25:37.277 delay adding those to the run, but your 00:25:37.360 --> 00:25:39.197 rate of producing checkpoints doesn't 00:25:39.280 --> 00:25:41.677 need to change. is just kind of the what 00:25:41.760 --> 00:25:45.517 those checkpoints represent could be 00:25:45.600 --> 00:25:48.957 less capable and specific axes more 00:25:49.040 --> 00:25:52.717 safety and security intertwined. But I would almost think of checkpoints and 00:25:52.800 --> 00:25:56.957 our shipping of models as a little bit more about taking a a photograph of 00:25:57.040 --> 00:26:00.077 where we happen to be, right? It's not like there's like a pre-ordained set of 00:26:00.160 --> 00:26:04.797 models that have to be created. It's we can co-design. And I think that in 00:26:04.880 --> 00:26:09.117 practice, I think that we actually have a lot of reasons that we prefer to do 00:26:09.200 --> 00:26:12.717 very incremental releases with these, you know, photographs taken with with a 00:26:12.800 --> 00:26:15.517 reasonable frequency. >> I guess what I'm getting at is that for 00:26:15.600 --> 00:26:19.277 a while, the release cadence felt relatively in place for everyone. In the 00:26:19.360 --> 00:26:22.557 last year, especially, it feels like it sped up. You know, new models coming out 00:26:22.640 --> 00:26:25.837 every six weeks essentially, you and Anthropic putting out new models 00:26:25.920 --> 00:26:30.397 constantly. Um, and what you're describing, this kind of taking stock 00:26:30.480 --> 00:26:34.397 and and introducing more friction into the process, it does feel like that just 00:26:34.480 --> 00:26:37.197 inevitably has to slow future momentum. 00:26:37.280 --> 00:26:40.237 And I'm I'm curious if you're thinking about that from the perspective of this 00:26:40.320 --> 00:26:43.997 machine we've been talking about, the business that you're now accountable for 00:26:44.080 --> 00:26:48.877 and the impacts on that or is it just is it too early to know? Look, I I just I 00:26:48.960 --> 00:26:52.157 think that like I understand that perspective and I think it's a very 00:26:52.240 --> 00:26:55.917 reasonable one, but I think that my observation of how we operate is that we 00:26:56.000 --> 00:27:00.077 co-design everything, right? So, it's like the way in which the business 00:27:00.160 --> 00:27:03.757 operates for sure is we will take the most capable model we are ready to 00:27:03.840 --> 00:27:06.157 release, right? We've done the appropriate safety and security controls 00:27:06.240 --> 00:27:10.797 for and we will bring that to our customers and help them get the ROI that 00:27:10.880 --> 00:27:14.477 they're looking for, help them really transform their businesses, get value. 00:27:14.560 --> 00:27:17.837 Simple in a lot of ways. And then the question of well what models are ready 00:27:17.920 --> 00:27:21.437 to release like we have many models that we produce internally and the choice of 00:27:21.520 --> 00:27:25.837 how high a bar to set how low a bar to set in terms of the capability jumps all 00:27:25.920 --> 00:27:29.837 of that has to be co-designed the frequency of it um that we have early 00:27:29.920 --> 00:27:34.477 access programs there's programs like daybreak red and blue which let us have 00:27:34.560 --> 00:27:37.677 different models that go out to to you know different partners and so I guess 00:27:37.760 --> 00:27:41.917 the way I would think about it is that that there is for sure a shift here in 00:27:42.000 --> 00:27:45.757 terms of our internal balance of of effort and where the checks checks and 00:27:45.840 --> 00:27:49.277 balances are and how how those things operate. I don't think that there is a 00:27:49.360 --> 00:27:51.757 clear trickle down to just like okay the 00:27:51.840 --> 00:27:53.997 model cadence will go from 6 weeks to 6 00:27:54.080 --> 00:27:57.677 months but the what those six weekly 00:27:57.760 --> 00:28:01.437 releases actually contain will be different. 00:28:01.520 --> 00:28:04.397 >> So maybe the the frontier the raw kind 00:28:04.480 --> 00:28:08.797 of frontier doesn't get pulled into production into the commercial world 00:28:08.880 --> 00:28:12.157 quite as fast even though it's still being trained on and worked on. 00:28:12.240 --> 00:28:13.597 >> Well I would think of this as an overall 00:28:13.680 --> 00:28:17.597 pacing, right? an overall pacing of of that frontier roll back. And another 00:28:17.680 --> 00:28:21.357 thing to think about is the fact that >> the 00:28:21.440 --> 00:28:25.997 our ability to produce these models is also and to get those capability gains 00:28:26.080 --> 00:28:30.317 is also increasing, right? Our raw capacity is because partly we understand 00:28:30.400 --> 00:28:33.277 this technology a lot better than we did in the past and that's actually very 00:28:33.360 --> 00:28:37.517 beneficial from a safety and alignment perspective too, right? That we really 00:28:37.600 --> 00:28:40.717 understand a lot of how the technology works. there's still many many pieces 00:28:40.800 --> 00:28:43.917 that we don't understand but our ability to engineer it and to really do a good 00:28:44.000 --> 00:28:48.637 job of of it technically um I think has has increased a lot um and that there 00:28:48.720 --> 00:28:52.717 are certain problems that really used to hold us back that are much diminished 00:28:52.800 --> 00:28:56.797 for example these big training runs you know they run for many months and so 00:28:56.880 --> 00:29:00.157 typically we'd launch them from a branch of code and so we'd be adding little 00:29:00.240 --> 00:29:04.877 fixes to that that are specific to that run at the same times can continue their 00:29:04.960 --> 00:29:07.757 mainline development and every so often there's something on the mainline that 00:29:07.840 --> 00:29:10.877 we're like oh we really want that change, we really want that fix. And 00:29:10.960 --> 00:29:13.357 then we're like, all right, we got to bring these two branches of development 00:29:13.440 --> 00:29:16.317 together and we merge them. And then we launch the run and we find that there's 00:29:16.400 --> 00:29:20.237 a bug because no one has actually tested all these mainline fixes at the massive 00:29:20.320 --> 00:29:23.517 scale because we only have one instance where we can do it. And then we spend 00:29:23.600 --> 00:29:26.717 the next three to four weeks with like all of our best people scouring trying 00:29:26.800 --> 00:29:30.317 to figure out where did the bug come from because you can't possibly read all 00:29:30.400 --> 00:29:31.597 of the code that's been written, right? 00:29:31.680 --> 00:29:35.917 No human could possibly do that. Today >> you run into that same type of issue. 00:29:36.000 --> 00:29:38.877 Codex just reads all the code, >> right? It finds the bug. it says here it 00:29:38.960 --> 00:29:40.957 is no problem. And so there's something 00:29:41.040 --> 00:29:45.917 about the partnership of this amplification of what the models can do 00:29:46.000 --> 00:29:48.957 that knock out whole classes of problems where we know it would take three to 00:29:49.040 --> 00:29:50.557 four weeks and now it takes a few hours. 00:29:50.640 --> 00:29:54.317 We've seen this concretely multiple times. And so what I'm getting at is 00:29:54.400 --> 00:29:59.117 that there are a lot of limitations for how we've had to develop in the past 00:29:59.200 --> 00:30:02.477 that have made it a lot harder for us to move forward on problems that are 00:30:02.560 --> 00:30:06.557 important. And now we have the opportunity to move forward on them in a 00:30:06.640 --> 00:30:09.437 much more efficient fashion. And that 00:30:09.520 --> 00:30:12.477 that to me is like let's use that extra 00:30:12.560 --> 00:30:16.797 juice to increase the alignment properties, to increase the safety 00:30:16.880 --> 00:30:19.837 properties. Let's make sure that we can have models that monitor other models in 00:30:19.920 --> 00:30:22.717 a trustworthy fashion. let's really implement some of these ideas that we 00:30:22.800 --> 00:30:25.197 were talking about back in 2017, right? 00:30:25.280 --> 00:30:27.757 That we have things like iterated 00:30:27.840 --> 00:30:32.317 amplification, things like debate where the models can really spend a bunch of 00:30:32.400 --> 00:30:36.717 compute in order to make sure that what they're doing is actually reasonable. To 00:30:36.800 --> 00:30:39.677 me, that's the opportunity in front of us. 00:30:39.760 --> 00:30:44.317 >> Does this these changes you're talking about impact the release of Astra, the 00:30:44.400 --> 00:30:47.117 thing that is coming sooner at all? Do you think that's going to be coming 00:30:47.200 --> 00:30:51.517 later than you guys originally hoped? Um well it is already later than we hoped 00:30:51.600 --> 00:30:54.797 in some ways. Yeah. Right. So we've had we've had a model that we think is very 00:30:54.880 --> 00:30:57.277 good. Lots of people using it internally. It's not out there yet. 00:30:57.360 --> 00:30:59.117 Yeah. >> And so I think that the answer is yes. 00:30:59.200 --> 00:31:01.757 We have actually been really like you 00:31:01.840 --> 00:31:03.837 know despite everything I just said we 00:31:03.920 --> 00:31:08.957 have very much sacrificed the shortterm maximalist like if you just wanted to 00:31:09.040 --> 00:31:12.557 Yeah. >> You know focus on revenue or something like that. 00:31:12.640 --> 00:31:15.997 >> That isn't how we've thought about it at all. No one has even as asked that 00:31:16.080 --> 00:31:19.437 question. Right. We've been really >> Yeah. Yeah. because it's just very clear 00:31:19.520 --> 00:31:20.957 like we know what our priorities are. 00:31:21.040 --> 00:31:25.277 Yeah. >> Like it is our priorities are AGI do right by AGI and our priority two is 00:31:25.360 --> 00:31:29.117 transform the economy. So it is on the priority list. Revenue is absolutely a 00:31:29.200 --> 00:31:32.157 priority for us. But we know the relative prioritization. We know what 00:31:32.240 --> 00:31:34.557 wins if there's a you know direct choice 00:31:34.640 --> 00:31:39.357 that has to be made. >> Do you think you have AGI? 00:31:39.440 --> 00:31:43.117 >> I think about this a lot. I talk to I talk to different people about where 00:31:43.200 --> 00:31:45.517 they believe we're at. And the best 00:31:45.600 --> 00:31:48.557 answer I've gotten from our team and I 00:31:48.640 --> 00:31:52.557 it really resonates for me is that AGI is a gradual thing, right? It's not 00:31:52.640 --> 00:31:55.517 really like a moment in time like a you know 00:31:55.600 --> 00:31:58.397 >> you talked about it like a moment in time at one point. The company 00:31:58.480 --> 00:32:00.077 definitely did. >> That's right. That's what we used to 00:32:00.160 --> 00:32:03.037 think, right? We we really when we set out we almost thought of it as nothing 00:32:03.120 --> 00:32:05.757 and then we'll put together the pieces and then they have AGI 00:32:05.840 --> 00:32:09.437 >> and I think what we've learned over the years is that was a naive perspective 00:32:09.520 --> 00:32:11.197 and you can see this in the evidence of 00:32:11.280 --> 00:32:14.717 like how smooth in many ways the 00:32:14.800 --> 00:32:18.877 creation of these agents has been how the transformation of software 00:32:18.960 --> 00:32:21.197 engineering has been now the transformation of knowledge work 00:32:21.280 --> 00:32:25.117 generally how cyber capabilities all these things they are gradually 00:32:25.200 --> 00:32:27.837 increasingly it's like one of these things where there are moments where you 00:32:27.920 --> 00:32:31.197 realize what has happened hugging face is a moment where you realize where you 00:32:31.280 --> 00:32:35.757 are, but it's all gradual. Like if you look back at kind of these different 00:32:35.840 --> 00:32:38.157 models that have been released over time and their performance on different 00:32:38.240 --> 00:32:40.877 benchmarks, it's just, you know, sort of 00:32:40.960 --> 00:32:44.397 that that exponential. And so I think 00:32:44.480 --> 00:32:48.157 that being said, the rest of the answer is that if you go if we go forward two 00:32:48.240 --> 00:32:52.717 years and we look back, I think that that we will say this is when AGI was 00:32:52.800 --> 00:32:54.877 created. So I think we are going through that moment, 00:32:54.960 --> 00:32:58.797 >> but it's not it's not something where you could just say it now. I don't think 00:32:58.880 --> 00:33:01.997 that it's ever going to be as simple as you can say this was the date. There's 00:33:02.080 --> 00:33:05.837 so many axes. And I think that maybe one lesson in general to the whole AGI 00:33:05.920 --> 00:33:07.677 conversation and why it's so complicated 00:33:07.760 --> 00:33:13.517 is because humans I think are capable of much more than a naive view would give 00:33:13.600 --> 00:33:16.157 credit to. >> I think that the diversity of what we 00:33:16.240 --> 00:33:19.277 do, how we set goals, how we accomplish 00:33:19.360 --> 00:33:23.757 goals, how tasks work, how jobs work, all these things extremely deep. And so 00:33:23.840 --> 00:33:25.517 I think that what we're seeing is that 00:33:25.600 --> 00:33:28.797 there's different aspects of either how 00:33:28.880 --> 00:33:33.837 people accomplish tasks or limitations that are hard for humans to overcome 00:33:33.920 --> 00:33:37.757 like the thing I mentioned with read all the code where these AIs are becoming 00:33:37.840 --> 00:33:42.397 able to fill in those gaps are able to do more and more and that humans it 00:33:42.480 --> 00:33:46.717 turns out that there are some things that we also really find are extremely 00:33:46.800 --> 00:33:48.637 important for it to be a human there. 00:33:48.720 --> 00:33:51.677 Like where does the judgment come from ultimately? Where does the taste come 00:33:51.760 --> 00:33:54.957 from? who decides what's good and what's not good. Like that is something that 00:33:55.040 --> 00:33:58.877 should be uniquely human. And I'll tell you just to give you an example. One 00:33:58.960 --> 00:34:02.957 thing we've seen with image generation, when you just generate an image with 00:34:03.040 --> 00:34:07.277 cashbt and you show it to someone, no one cares. It is uninteresting to anyone 00:34:07.360 --> 00:34:09.437 besides you. But if you take a photo of 00:34:09.520 --> 00:34:11.517 your family and you say, "Turn this into 00:34:11.600 --> 00:34:14.637 the into a caricature, turn this into, 00:34:14.720 --> 00:34:18.877 you know, a cartoon style," people love it. Goes viral. And so there's something 00:34:18.960 --> 00:34:23.117 about that element of humanity that matters. so deeply people want to 00:34:23.200 --> 00:34:27.757 connect with other people and that is actually in in my mind the purpose of 00:34:27.840 --> 00:34:31.757 this technology a thing that I think is is really emerging is for deepening 00:34:31.840 --> 00:34:35.197 human connection so you can spend more time with your family you can spend more 00:34:35.280 --> 00:34:38.397 time with people you want to spend time with less time you know typing away at a 00:34:38.480 --> 00:34:40.477 machine less time you know getting 00:34:40.560 --> 00:34:43.517 carpal tunnel the whole thing and so I 00:34:43.600 --> 00:34:45.997 think that there is something about what 00:34:46.080 --> 00:34:50.877 AGI is as it's starting to unfold it's almost this like fog that's lifting it's 00:34:50.960 --> 00:34:54.397 almost teaching us more about ourselves and what it means to be human. 00:34:54.480 --> 00:34:58.077 >> Given the capabilities that you guys have internally and the measures you're 00:34:58.160 --> 00:34:59.997 taking now, moving compute to alignment 00:35:00.080 --> 00:35:02.797 and putting in all these checkpoints, 00:35:02.880 --> 00:35:05.197 um, how are you going to know when 00:35:05.280 --> 00:35:07.437 you're in a good spot there and and that 00:35:07.520 --> 00:35:11.037 you've solved the problems and the 00:35:11.120 --> 00:35:15.677 models are aligned? So I think the perspective here is is very critical and 00:35:15.760 --> 00:35:20.317 this is something we've again spent lots of time over the years thinking about 00:35:20.400 --> 00:35:25.517 debating really trying to refine a perspective. In my view it really comes 00:35:25.600 --> 00:35:28.717 down to iteration. It comes down to iterative deployment iterative 00:35:28.800 --> 00:35:32.157 development. So you need to have checks and balances. You need to have 00:35:32.240 --> 00:35:34.797 evaluations. You need to have bars. You need to have all of these things. You 00:35:34.880 --> 00:35:39.197 need to have governance. You need to have a setup that is technically sound. 00:35:39.280 --> 00:35:42.957 And the way to think about it is everything is gradual. Everything is 00:35:43.040 --> 00:35:45.677 incremental and you learn and you 00:35:45.760 --> 00:35:49.997 update, you iterate. And that as the model capabilities increase, the 00:35:50.080 --> 00:35:53.597 tightness of those bounds has to increase correspondingly. And so think 00:35:53.680 --> 00:35:56.397 of it a little bit like if you're launching a rocket, right? If the 00:35:56.480 --> 00:35:59.437 rocket's very small, like there's a lot of things you can probably do wrong and 00:35:59.520 --> 00:36:03.117 it's fine, right? You just kind of like launch it and like, you know, it's it's 00:36:03.200 --> 00:36:07.517 all good. Rocket bigger, the tolerance is much more tiny. you know, you go 00:36:07.600 --> 00:36:11.117 from, you know, multiple centimeters to m multiple millimeters. That's where 00:36:11.200 --> 00:36:15.357 where you have to be. And as you start to add more responsibility to what's 00:36:15.440 --> 00:36:18.077 what the payloads are for the rocket, where it's going, what it's doing, you 00:36:18.160 --> 00:36:21.997 think about all the the things that are relying on that rocket and how all the 00:36:22.080 --> 00:36:25.197 systems that have to go right, all the things that go wrong. You have to 00:36:25.280 --> 00:36:29.517 continue to ratchet your ability to be able to measure and to ensure that 00:36:29.600 --> 00:36:32.957 you're able to even do the engineering in addition to the safety and security 00:36:33.040 --> 00:36:37.197 properties of it. And that is very much how we think about it is that there are 00:36:37.280 --> 00:36:41.357 moments where we do introduce very punctuated change. When we made the 00:36:41.440 --> 00:36:44.397 preparedness framework the first time was an example of that. But even the 00:36:44.480 --> 00:36:48.237 preparedness framework we very carefully very deliberately baked in the ability 00:36:48.320 --> 00:36:51.597 to change it. We said there will be new information. We will have to learn. We 00:36:51.680 --> 00:36:55.837 will have to update. But at any point we wanted it to be clear what are our bars. 00:36:55.920 --> 00:36:59.597 How do we handle it when a model reaches a certain level of capability? And we 00:36:59.680 --> 00:37:01.037 recently recently talked about the fact 00:37:01.120 --> 00:37:03.917 that Astra is approaching cyber 00:37:04.000 --> 00:37:08.557 critical. It's not necessarily there because 00:37:08.640 --> 00:37:10.317 in order to really prove it requires 00:37:10.400 --> 00:37:12.797 eval aren't fully that you know that 00:37:12.880 --> 00:37:17.757 aren't fully defined but that we are saying okay it's starting to approach it 00:37:17.840 --> 00:37:21.037 so let's treat it as if it is. And so then that means we need to go and 00:37:21.120 --> 00:37:24.317 execute on certain safeguards and certain checks and balances. And you're 00:37:24.400 --> 00:37:27.597 seeing all of that play out. And so I think that the answer to your question 00:37:27.680 --> 00:37:30.717 is that we define a bar, we hit the bar. 00:37:30.800 --> 00:37:33.997 You can't you have to take a framework that looks that way, right? To really 00:37:34.080 --> 00:37:37.277 focus on the iteration. Are you learning? And are the the bars that 00:37:37.360 --> 00:37:39.837 you're setting tightening in a 00:37:39.920 --> 00:37:44.397 sufficiently sort of in a in a sufficiently executed way in order to 00:37:44.480 --> 00:37:45.837 deliver the properties that are required. 00:37:45.920 --> 00:37:50.317 >> Yeah. I mean your rocket analogy is an interesting one because um rockets also 00:37:50.400 --> 00:37:52.637 explode a lot and I've been thinking 00:37:52.720 --> 00:37:55.997 about uh your all's concern as these 00:37:56.080 --> 00:37:58.797 things get more capable yeah the concern 00:37:58.880 --> 00:38:03.357 of controlling them and how you can and whether there's kind of like a Pandora's 00:38:03.440 --> 00:38:06.237 box situation that's unfolding right now and I don't know if you've thought about 00:38:06.320 --> 00:38:09.917 that personally recently um with the insights you guys have on even what's 00:38:10.000 --> 00:38:12.077 ahead of Astra and what's causing you to 00:38:12.160 --> 00:38:15.037 to make this slowdown um but yeah do you 00:38:15.120 --> 00:38:19.437 do you worry about We view it as our job to really think through the implications 00:38:19.520 --> 00:38:22.557 of this technology. Like that is why we're here is that we want to help steer 00:38:22.640 --> 00:38:25.597 it in a more positive direction that would without us. Like that that that is 00:38:25.680 --> 00:38:26.877 that is table stakes to our mission. 00:38:26.960 --> 00:38:28.957 That is how we approach things. And I 00:38:29.040 --> 00:38:32.077 think it is important to 00:38:32.160 --> 00:38:34.237 take two two lenses to what you just 00:38:34.320 --> 00:38:39.437 said. The first one is that again this technology is not being created in a 00:38:39.520 --> 00:38:43.037 vacuum, right? that there are many people, many actors, many companies, 00:38:43.120 --> 00:38:45.197 many organizations, many countries even 00:38:45.280 --> 00:38:47.037 that are pursuing the creation of this 00:38:47.120 --> 00:38:50.077 technology. And we happen to be in a 00:38:50.160 --> 00:38:52.957 spot where we have some lead, we have 00:38:53.040 --> 00:38:55.357 some ability to see into that future. 00:38:55.440 --> 00:38:58.397 And it is both important to think about 00:38:58.480 --> 00:39:02.477 how do we develop responsibly, safely, and securely. That's something that we 00:39:02.560 --> 00:39:05.997 spend a lot of time on. You can see it very concretely how we prioritize that 00:39:06.080 --> 00:39:09.357 in everything we do. And I think that actually mistakes we made in the past is 00:39:09.440 --> 00:39:13.917 actually talking less about the fact that this is how we operate that safety 00:39:14.000 --> 00:39:18.077 is such a first class priority first class requirement 00:39:18.160 --> 00:39:22.077 just doing it but not necessarily talking about it as publicly and I think 00:39:22.160 --> 00:39:25.357 that that's something we're starting to realize that we do need to help people 00:39:25.440 --> 00:39:29.437 understand that that is how that is a core value of ours that is something we 00:39:29.520 --> 00:39:32.877 care about so deeply like I think we've always been the field leader in safety 00:39:32.960 --> 00:39:35.917 security alignment u but I don't think people have necessarily known that 00:39:36.000 --> 00:39:39.117 >> yeah even with the noisethropic makes about how they're the leader in that 00:39:39.200 --> 00:39:41.037 capacity. You think you guys have been? 00:39:41.120 --> 00:39:43.757 >> I believe at a substantive level that we have been. 00:39:43.840 --> 00:39:47.197 >> And do you think that what you guys are doing and the capabilities you're seeing 00:39:47.280 --> 00:39:49.757 and needing to redo, you know, the preparedness framework and all that 00:39:49.840 --> 00:39:53.917 implies you are actually just way ahead of maybe where everyone thought you were 00:39:54.000 --> 00:39:57.277 relative to anthropic and the other frontier labs? Like does this suggest 00:39:57.360 --> 00:39:59.597 that you all have a breakthrough 00:39:59.680 --> 00:40:03.037 internally that the world hasn't seen? 00:40:03.120 --> 00:40:06.797 Well, I can't comment on how, you know, 00:40:06.880 --> 00:40:10.157 on competitive positioning perhaps, but I guess I would just say that in my 00:40:10.240 --> 00:40:13.917 mind, we're very focused on the work that we're doing and that we think that 00:40:14.000 --> 00:40:18.077 it we see smooth scaling and I think that the exponential continues and so 00:40:18.160 --> 00:40:20.557 it's a little bit less about where are we relative to others, it's a little bit 00:40:20.640 --> 00:40:23.757 more about the way we look at it is where are we relative to expectation or 00:40:23.840 --> 00:40:27.997 what do we think we will be able to accomplish and I think that the reason 00:40:28.080 --> 00:40:29.917 this matters is that the bar that we 00:40:30.000 --> 00:40:32.877 hold ourselves to doesn't really change 00:40:32.960 --> 00:40:37.437 depending on kind of what the ecosystem is showing because there's so much noise 00:40:37.520 --> 00:40:41.037 out there. It's hard to know what's real that what we focus on is like are we 00:40:41.120 --> 00:40:45.277 doing right by the our own conscious by our own standards by what we believe is 00:40:45.360 --> 00:40:49.037 important and that is the number one lens that we take. And the second thing 00:40:49.120 --> 00:40:53.757 I wanted to add is that I think that the we you know there's this recent picture 00:40:53.840 --> 00:40:57.437 that came out of or this this open letter uh called pacing the frontier. We 00:40:57.520 --> 00:41:00.877 talked about how we support that and the the the language in there is very 00:41:00.960 --> 00:41:05.197 deliberately chosen. Pacing is a very important word, right? That if you think 00:41:05.280 --> 00:41:09.437 about what's happening is that what we're saying is that you can't take a 00:41:09.520 --> 00:41:12.797 just like one-dimensional view to what it means to deliver a model, right? If 00:41:12.880 --> 00:41:15.757 you just look at capability, don't look at alignment, you don't look at the kind 00:41:15.840 --> 00:41:19.037 of safety properties, security properties of how it was trained, it's 00:41:19.120 --> 00:41:23.197 too myopic. At this level of capability, you need to think about those as first 00:41:23.280 --> 00:41:26.637 class requirements. And in fact, they are the bottleneck, right? because those 00:41:26.720 --> 00:41:30.397 are the areas that I think people have sort of are you know have just focused 00:41:30.480 --> 00:41:34.477 on less in an integrated fashion and that integration by the way critical 00:41:34.560 --> 00:41:37.277 right that I think that where people go wrong is when they think that you can 00:41:37.360 --> 00:41:40.157 separate out these properties you can have just a safety you can have just a 00:41:40.240 --> 00:41:42.477 capability no it's one thing we're 00:41:42.560 --> 00:41:44.637 trying to deliver one system one AGI 00:41:44.720 --> 00:41:49.517 that uplifts you that is helpful to you that delivers on your goals that you can 00:41:49.600 --> 00:41:53.357 steer that is aligned with what you want like that is what we are delivering and 00:41:53.440 --> 00:41:57.357 so that back flows to all these questions of bars and testing and 00:41:57.440 --> 00:41:59.917 evaluations, safety, security. Um, all 00:42:00.000 --> 00:42:02.077 of that is is very very critical. 00:42:02.160 --> 00:42:06.477 >> I think I'll I'll pivot us a little bit because I want to hear you talk about um 00:42:06.560 --> 00:42:08.957 the merge and um what's going on with 00:42:09.040 --> 00:42:13.517 that. So, we've been talking about the existential risk, you know, what's 00:42:13.600 --> 00:42:16.957 happening at the frontier, which I think most people in the world don't see, 00:42:17.040 --> 00:42:21.917 right? That's happening inside companies like OpenAI. What people do see is how 00:42:22.000 --> 00:42:25.837 chat is evolving, how CodeEx is evolving. And one of your big priorities 00:42:25.920 --> 00:42:29.117 in this role, as I understand it, is what you guys call the merge internally, 00:42:29.200 --> 00:42:31.277 which is creating the super app, this 00:42:31.360 --> 00:42:34.157 combined chat GPT and codeex. Um, and 00:42:34.240 --> 00:42:39.357 you've got chat GPT for work now and and the tabs. Um, but at the same time, like 00:42:39.440 --> 00:42:42.637 it doesn't feel like that's the end state, you know, and I'm a user of it 00:42:42.720 --> 00:42:46.637 and I see brutalness in how it's currently executed. Some things don't 00:42:46.720 --> 00:42:50.397 sync. You can tell you're still combining text stacks. And I'm curious 00:42:50.480 --> 00:42:53.997 once you're done with that work, what is the end product for people that you 00:42:54.080 --> 00:42:58.397 imagine? >> A personal AGI. >> Personal AGI. And what how does that 00:42:58.480 --> 00:43:01.757 manifest as software? >> Yeah, it's 00:43:01.840 --> 00:43:05.837 it's it's a really really exciting vision in my mind, right? Because where 00:43:05.920 --> 00:43:09.837 we started with chatbt, it was a text box. You could ask anything. 00:43:09.920 --> 00:43:13.517 >> But if you think about it, it was limited in a very fundamental way. The 00:43:13.600 --> 00:43:17.437 only thing it could do is output some text for you. the only impact it could 00:43:17.520 --> 00:43:21.517 have in the world was you know kind of in your brain right that it wasn't able 00:43:21.600 --> 00:43:23.837 to look at your calendar it wasn't able 00:43:23.920 --> 00:43:26.877 to look at your email and all of that is 00:43:26.960 --> 00:43:30.717 now changed chatb work is really by the 00:43:30.800 --> 00:43:34.717 way a codeex harness in the cloud right so we had chat that was this you know 00:43:34.800 --> 00:43:38.637 question answer system um with iterated turns you know people actually then go 00:43:38.720 --> 00:43:42.157 beyond just you know quick questions to you know health like all these kinds of 00:43:42.240 --> 00:43:45.917 queries that are able to help them >> and we had codeex which was an agentic 00:43:46.000 --> 00:43:50.717 system that can go do work and it was so obvious. It's like chat has to become 00:43:50.800 --> 00:43:53.517 agentic as the place we started at the beginning of the year was to say well 00:43:53.600 --> 00:43:57.757 let's let's have great chat let's you know make it also do agent stuff and it 00:43:57.840 --> 00:44:01.917 just turned out to be so hard to execute because now we have two agentic stacks 00:44:02.000 --> 00:44:04.717 and then we compare between one and the other and there'd be something that work 00:44:04.800 --> 00:44:08.397 in this one we like why isn't working in that one and it was just so slow and so 00:44:08.480 --> 00:44:11.917 the decision we made instead is say let's take the agentic stack we have the 00:44:12.000 --> 00:44:14.797 codec stack and bring it to chat GBT and 00:44:14.880 --> 00:44:19.277 so what you see with chat work is exactly right this is an intermediate 00:44:19.360 --> 00:44:23.517 And in some ways, it's kind of funny if you think about it. ChatBT was a text 00:44:23.600 --> 00:44:25.757 box. CHATBT worked as a text box. 00:44:25.840 --> 00:44:27.917 >> But this text box is way more awesome. 00:44:28.000 --> 00:44:32.157 Way more. Exactly. And so where we want to be is you should 00:44:32.240 --> 00:44:33.837 >> it should be simple. The user should not 00:44:33.920 --> 00:44:36.397 have to think about this. You shouldn't have to think about tabs or switching 00:44:36.480 --> 00:44:39.277 and shouldn't have model pickers. You shouldn't have thinking levels. You 00:44:39.360 --> 00:44:42.477 should have an AGI. You should be able to ask it. And in fact, you shouldn't 00:44:42.560 --> 00:44:46.157 even have to ask. It should proactively say, "Hey, I noticed that your favorite 00:44:46.240 --> 00:44:50.717 band, because it knows you well, is in town. Tickets went on sale. I got you 00:44:50.800 --> 00:44:54.957 the tickets exactly where you want for you and your family, and um you know, I 00:44:55.040 --> 00:44:57.517 only had 10 minutes to do it because everything is going to go so fast. So, I 00:44:57.600 --> 00:45:00.397 just purchased it proactively. Hope that's okay." And if you've built enough 00:45:00.480 --> 00:45:03.677 trust with the agent, so you've given it permission to do that, and it kind of 00:45:03.760 --> 00:45:05.917 knows that, well, these tickets are within your your price range, within 00:45:06.000 --> 00:45:09.117 your budget, these other ones. Oh, no, no, I I got to get permission for that, 00:45:09.200 --> 00:45:14.157 right? That's the kind of system that we want. So trust is the key word. EQ also 00:45:14.240 --> 00:45:17.117 very important that if you've shared some information like you've talked 00:45:17.200 --> 00:45:20.557 about a health condition, maybe that's okay information for the AI to talk 00:45:20.640 --> 00:45:22.157 about with a parent, but maybe it's not 00:45:22.240 --> 00:45:24.077 okay to talk about with a stranger, 00:45:24.160 --> 00:45:29.517 right? All of that judgment and and also observability and really helping you 00:45:29.600 --> 00:45:33.677 build up that trust relationship with the AI so that you know what you know 00:45:33.760 --> 00:45:36.957 when it's operating on your behalf that it'll pursue your goals. Um that's very 00:45:37.040 --> 00:45:41.517 very critical. But the same thing same kind of technology the underlying AGI 00:45:41.600 --> 00:45:45.037 stack one stack one way of doing connectors so it's able to hook up to 00:45:45.120 --> 00:45:48.637 this context also applies to the enterprise and we're really seeing it 00:45:48.720 --> 00:45:51.917 not just for software engineering which by the way has been totally transformed 00:45:52.000 --> 00:45:55.277 this year and to the point of AGI being a gradual thing or a thing where we just 00:45:55.360 --> 00:45:59.357 constantly update our expectations like we all kind of forget what software was 00:45:59.440 --> 00:46:03.277 one year ago totally different >> very interesting thing within open AI is 00:46:03.360 --> 00:46:06.957 that we're at basically 100% adoption of chat GBT work across every function 00:46:07.040 --> 00:46:10.637 right sales finance everyone and whenever you know we roll 00:46:10.720 --> 00:46:13.597 out a new model and it goes away or you know the system breaks for whatever 00:46:13.680 --> 00:46:18.317 internal reason um everyone's so upset they're I cannot do my work and you're 00:46:18.400 --> 00:46:20.957 like I don't get it six months ago you didn't have this and you're doing your 00:46:21.040 --> 00:46:24.717 work just fine like why are you you know why are you complaining and you realize 00:46:24.800 --> 00:46:29.357 people adapt so quickly and learn how to get the most out of these tools and once 00:46:29.440 --> 00:46:32.797 you're there you also just sort of forget that it was any other way and so 00:46:32.880 --> 00:46:36.237 there's something about this where I think we will have these incredible 00:46:36.320 --> 00:46:40.157 amazing tools these is assistance that will remove so much of the toil that 00:46:40.240 --> 00:46:42.717 will mean that you can spend your time doing what you want to do. That you can 00:46:42.800 --> 00:46:46.557 focus on building a better life, you can focus on if you want to be making more 00:46:46.640 --> 00:46:48.957 money, if you want to be relaxing more, if you want to spend more time with your 00:46:49.040 --> 00:46:52.157 family, any of those things, you will have the agency to do it. You have the 00:46:52.240 --> 00:46:55.357 agency to build. If you have a vision, you have creativity, you want it to come 00:46:55.440 --> 00:46:57.597 into the world, you will have the tools 00:46:57.680 --> 00:47:02.877 to do that. So, I think that what chat GBT will do is help you with what you 00:47:02.960 --> 00:47:06.077 want to accomplish. If you want to make it your world better, that is what chat 00:47:06.160 --> 00:47:08.557 will help you accomplish. >> And the implications of that are that 00:47:08.640 --> 00:47:11.197 what the coding AI tools have done to 00:47:11.280 --> 00:47:15.997 engineering this could do to all knowledge work, all kinds of fields. And 00:47:16.080 --> 00:47:19.917 that's how you guys used to talk about AGI is like, you know, it was basically 00:47:20.000 --> 00:47:23.437 AI that could do most economically viable work better than humans. And what 00:47:23.520 --> 00:47:26.557 you're describing now in the product kind of sounds like that. 00:47:26.640 --> 00:47:29.677 >> I think that the amplification we see in software engineering, we expect across 00:47:29.760 --> 00:47:34.477 every field for sure. And there is something that I think that we is also 00:47:34.560 --> 00:47:36.477 worth really thinking about which is the 00:47:36.560 --> 00:47:39.277 software engineers rather than having 00:47:39.360 --> 00:47:43.677 more vacation time and having more chill job they're all working harder than ever 00:47:43.760 --> 00:47:45.677 before. >> Yeah. You guys work a lot of hours here. 00:47:45.760 --> 00:47:49.197 >> Yes. And I think across the whole industry people are talking about that 00:47:49.280 --> 00:47:53.437 because now you feel like if your agents aren't running it's just wasted 00:47:53.520 --> 00:47:57.437 opportunity. Right. It's like and I know that feeling because we have spent years 00:47:57.520 --> 00:48:00.557 of we have these GPU clusters. you're either running something on them or 00:48:00.640 --> 00:48:03.677 you've lost the opportunity. You can't get that time back. It's not like the 00:48:03.760 --> 00:48:07.117 GPU hours that you're not using while you're asleep overnight. You can bank 00:48:07.200 --> 00:48:11.037 them up and use them later. It's use it or lose it. And that is very much the 00:48:11.120 --> 00:48:13.037 opportunity that everyone has in front 00:48:13.120 --> 00:48:15.917 of them is you have almost an AGI, maybe 00:48:16.000 --> 00:48:20.877 soon truly an AGI in your pocket. Can do anything you want. What is it you want? 00:48:20.960 --> 00:48:23.597 It's >> waiting on you. It's waiting on your direction, your guidance, your 00:48:23.680 --> 00:48:25.597 management. >> It would be nice if eventually we can 00:48:25.680 --> 00:48:28.717 get some time back from AI. >> That's right. So like right now I do 00:48:28.800 --> 00:48:30.877 feel like it it will help you work more. 00:48:30.960 --> 00:48:34.717 >> That's right. And so I think that I think that that is that is exactly a 00:48:34.800 --> 00:48:38.477 next step, right? Is really moving from this like you're micromanaging this 00:48:38.560 --> 00:48:41.197 agent. You have to provide it with these tasks and you have to give the context 00:48:41.280 --> 00:48:42.877 because it just doesn't have it steer. 00:48:42.960 --> 00:48:46.397 >> Exactly. You want something that has much more like you think about like a 00:48:46.480 --> 00:48:50.957 really good assistant who really is plugged in and you've given it again you 00:48:51.040 --> 00:48:52.157 have to build up trust with it, right? 00:48:52.240 --> 00:48:55.197 You have to really say like okay I trust you with this context. I trust you with 00:48:55.280 --> 00:48:58.957 this responsibility. I trust you with this job. whatever it is. And once it's 00:48:59.040 --> 00:49:02.477 there, it should be able to proactively run and really execute. You wake up in 00:49:02.560 --> 00:49:06.157 the morning, you know, say, "Hey, here are five things I need from you to 00:49:06.240 --> 00:49:08.637 unblock me. You know, I just need approval on this. Like, here's a 00:49:08.720 --> 00:49:09.997 question I have." Those kinds of things. 00:49:10.080 --> 00:49:13.037 As you're drinking your morning coffee, you say yes, no, yes, no. 00:49:13.120 --> 00:49:16.637 >> Then you go come back, you know, late at night once again, you've got five more 00:49:16.720 --> 00:49:20.237 things. And so, I think that it's going to be a world where it's going to be 00:49:20.320 --> 00:49:23.757 different from, I think, the kind of AGI world that many of us would have 00:49:23.840 --> 00:49:26.477 pictured 5 10 years ago, right? It's not 00:49:26.560 --> 00:49:29.037 like I think that this integration of 00:49:29.120 --> 00:49:34.237 humans really being maximally empowered with this tool, but you still have 00:49:34.320 --> 00:49:37.117 involvement, right? That you still have oversight, you still have judgment, you 00:49:37.200 --> 00:49:39.757 still have goals. And also on the flip side, I think you still have 00:49:39.840 --> 00:49:43.917 accountability, right? The question of is, you know, if something doesn't get 00:49:44.000 --> 00:49:47.837 done, whose fault is it? I don't think that blaming JBT saying, "Oh, it was too 00:49:47.920 --> 00:49:52.077 lazy, didn't do the thing," is going to fly, right? I think that you as the 00:49:52.160 --> 00:49:55.677 director, as the person who's who's sort of running the show, you know that you 00:49:55.760 --> 00:49:58.477 need to really feel that accountability, make sure the task gets done the way 00:49:58.560 --> 00:50:00.957 that you want. So, you have this incredible tool, incredible 00:50:01.040 --> 00:50:03.997 amplification, but just like >> it's almost like everyone's going to 00:50:04.080 --> 00:50:07.277 become the CEO of this corporation of as 00:50:07.360 --> 00:50:11.597 many agents as you as you care to orchestrate. Um, it's going to be just 00:50:11.680 --> 00:50:16.157 like a very empowering world where people can do so much more of they want 00:50:16.240 --> 00:50:19.757 what they want to do. Um, and I think that it'll be a world where people will 00:50:19.840 --> 00:50:23.757 be able to make more money. I think that people will be able to improve their 00:50:23.840 --> 00:50:26.957 health. I think you'll be able to learn anything you want to learn and you'll be 00:50:27.040 --> 00:50:31.597 able to spend your time the way that you want to. And I think all of that means 00:50:31.680 --> 00:50:35.197 that if you have ambitions that you want to realize that they're more insight 00:50:35.280 --> 00:50:38.637 than ever. And I I know this very concretely from talking to different 00:50:38.720 --> 00:50:42.477 users, different customers, people who run small businesses on top of chatbt 00:50:42.560 --> 00:50:46.077 and say, "I would not have been able to do this otherwise." like people who 00:50:46.160 --> 00:50:50.157 actually quit their job to pursue a business because they said hey I could 00:50:50.240 --> 00:50:55.037 do this now because chat can help me with so much of it and that kind of 00:50:55.120 --> 00:50:58.237 empowerment where people wouldn't have been able to take the leap before but 00:50:58.320 --> 00:51:02.077 now can I think that that is it's like the American dream on steroids like that 00:51:02.160 --> 00:51:04.157 is what we're unlocking >> and it's I mean you're you're painting 00:51:04.240 --> 00:51:08.077 such a rosy picture and I do see that too even in my own work as like a you 00:51:08.160 --> 00:51:11.837 know solo media entrepreneur and we can land it here but I also you know people 00:51:11.920 --> 00:51:15.357 feel very negatively about AI especially in the United States as And I'm 00:51:15.440 --> 00:51:18.877 wondering if like the stakes are actually you have to deliver on the 00:51:18.960 --> 00:51:22.557 product you're describing that will empower people this way for that to 00:51:22.640 --> 00:51:25.677 shift. Maybe that is the thing that causes sentiment to shift because right 00:51:25.760 --> 00:51:28.397 now it feels increasingly negative in a 00:51:28.480 --> 00:51:33.357 lot of corners of society. Uh and I'm wor I'm curious like as the person you 00:51:33.440 --> 00:51:37.437 know responsible now for for the business and chatbt and all of that do 00:51:37.520 --> 00:51:39.517 you feel that do you feel that connection? 00:51:39.600 --> 00:51:42.397 >> We we feel it. We feel the responsibility. We feel the gravity and 00:51:42.480 --> 00:51:46.877 we definitely feel the criticality of both us delivering on this vision and 00:51:46.960 --> 00:51:49.917 helping people understand it. Helping people understand who we are 00:51:50.000 --> 00:51:53.277 >> better than we have in the past and help them understand our intentions, why 00:51:53.360 --> 00:51:57.437 we're doing this. And I do think that there's a different set of sentiments 00:51:57.520 --> 00:51:58.717 depending on where you go in the world. 00:51:58.800 --> 00:52:02.717 Yeah. Right. That it's very clear in America like we understand that we have 00:52:02.800 --> 00:52:06.317 a lot of uh you know we need to deliver on this this vision to really help 00:52:06.400 --> 00:52:11.277 people lift themselves up to really empower people. um countries like Korea, 00:52:11.360 --> 00:52:14.477 Japan, I think that actually sentiment is very positive towards these tools. 00:52:14.560 --> 00:52:19.037 People really see that potential and I think that where America sits right now 00:52:19.120 --> 00:52:23.117 is at an incredible moment with incredible opportunity and position. 00:52:23.200 --> 00:52:28.077 Like we are the leader in this technology. This is going to be the most 00:52:28.160 --> 00:52:29.837 important technology that humanity ever 00:52:29.920 --> 00:52:31.837 creates, right? This is going to be this 00:52:31.920 --> 00:52:36.077 technology that will really shape how 00:52:36.160 --> 00:52:40.317 people live, how people work, how people accomplish goals, how economy, how the 00:52:40.400 --> 00:52:44.477 economy runs, how countries relate to each other, all of these things. And 00:52:44.560 --> 00:52:49.597 America leads in that. And I think that that opportunity is one that we should 00:52:49.680 --> 00:52:53.517 not take for granted. We should take make the most of it. And so that has 00:52:53.600 --> 00:52:57.437 been one of the biggest realizations for me over the course of OpenAI is to 00:52:57.520 --> 00:53:01.117 realize that this is going to be so important and it has to be something 00:53:01.200 --> 00:53:03.437 that is broadly empowering, right? It 00:53:03.520 --> 00:53:06.797 has to be a technology that's available 00:53:06.880 --> 00:53:11.277 to everyone to lift themselves up. But America leading and ensuring that the 00:53:11.360 --> 00:53:14.637 values that are represented by this technology, the way that it plays out in 00:53:14.720 --> 00:53:16.877 the world being something that enhances 00:53:16.960 --> 00:53:18.797 democratic values that really does 00:53:18.880 --> 00:53:23.037 uplift all of those those um you know 00:53:23.120 --> 00:53:27.357 the core principles that we hold dear that is at the core and so I think that 00:53:27.440 --> 00:53:31.597 it's about how we use it but also about how do we use this leadership 00:53:31.680 --> 00:53:33.917 opportunity and how do we ensure we don't squander it. 00:53:34.000 --> 00:53:36.797 >> Yeah, those are all my questions. Do you 00:53:36.880 --> 00:53:40.157 feel like there's anything on your mind that you want to say or anything that 00:53:40.240 --> 00:53:42.077 we're missing? >> So many things. 00:53:42.160 --> 00:53:44.157 >> There's a lot of things. >> There's so much ground to cover. I think 00:53:44.240 --> 00:53:47.837 at the end of the day, what we are doing 00:53:47.920 --> 00:53:50.557 is in some ways incredibly complex. 00:53:50.640 --> 00:53:53.117 There's so many facets to what we're building. You think about consumer and 00:53:53.200 --> 00:53:56.077 enterprise and the fact we're bringing these things together and that what an 00:53:56.160 --> 00:53:59.117 enterprise is is going to change because people are going to have this unlocked 00:53:59.200 --> 00:54:02.717 wave of entrepreneurship. All that's real. We have this research program. The 00:54:02.800 --> 00:54:06.157 research isn't just about we're we're off writing papers. It's like people who 00:54:06.240 --> 00:54:10.557 really care about how their technology will be used, making sure the safety and 00:54:10.640 --> 00:54:14.397 alignment that uh slides, doc, spreadsheet, all of those are good. It's 00:54:14.480 --> 00:54:19.037 just like all of this complexity, but at the core, there's something very simple. 00:54:19.120 --> 00:54:23.597 We have a mission. Our mission is to ensure artificial general intelligence 00:54:23.680 --> 00:54:27.097 benefits all of humanity. That's what we're here to do. Everything we do flows 00:54:27.180 --> 00:54:30.637 [music] from that one singular goal. So that's what we're pursuing. That's why 00:54:30.720 --> 00:54:35.800 we started this place and that is what we're committed to.