WEBVTT 00:00:00.000 --> 00:00:01.917 OpenAI president and co-founder Greg 00:00:02.000 --> 00:00:04.077 Brockman joins us to discuss OpenAI's 00:00:04.160 --> 00:00:08.277 newest model, Spud, aka GPT 5.5, [music] 00:00:08.360 --> 00:00:11.757 and where it leaves OpenAI competitively. That's coming up right 00:00:11.840 --> 00:00:15.517 after this. Open the Big Technology podcast, today we have an emergency 00:00:15.600 --> 00:00:17.837 episode with OpenAI president and 00:00:17.920 --> 00:00:20.117 co-founder Greg Brockman all about GPT 00:00:20.200 --> 00:00:25.077 5.5, the famous Spud model. Looking at what it does and what it means for 00:00:25.160 --> 00:00:27.917 OpenAI. Greg, great to see you. Welcome back to the show. 00:00:28.000 --> 00:00:31.477 >> Thank you for having me. Hope it's not too much of an emergency. 00:00:31.560 --> 00:00:36.757 >> Well, I am definitely recording in in a Vegas hotel room, so 00:00:36.840 --> 00:00:39.957 more emergency than our last conversation, but we had some time to 00:00:40.040 --> 00:00:45.037 prepare. So, it's great to be on with you. So, let's let's just start with 00:00:45.120 --> 00:00:49.837 with this Can you confirm GPT 5.5 is Spud? 00:00:49.920 --> 00:00:52.397 >> Yes. >> Okay. 00:00:52.480 --> 00:00:56.037 What is GPT 5.5? 00:00:56.120 --> 00:00:57.997 >> Well, it's it's an amazing model. I 00:00:58.080 --> 00:01:01.477 think in many ways it is a step towards 00:01:01.560 --> 00:01:05.197 a new way of getting work done with a computer. It's a new class of 00:01:05.280 --> 00:01:09.797 intelligence. It's extremely useful at things like programming, right? And all 00:01:09.880 --> 00:01:14.277 the different aspects of debugging and solving very hard and gnarly problems, 00:01:14.360 --> 00:01:18.077 just being very proactive, and really being able to solve problems end to end 00:01:18.160 --> 00:01:22.797 with little instruction. But, the thing that's to me most remarkable is not 00:01:22.880 --> 00:01:26.877 necessarily the fact that it got better at coding. Like that, I think is is what 00:01:26.960 --> 00:01:29.637 everyone kind of expects. But, the fact that it's now really crossed the 00:01:29.720 --> 00:01:33.877 threshold of usefulness for general kinds of applications. And so, it's much 00:01:33.960 --> 00:01:38.317 better at creating slides, spreadsheets, much better at computer use, using your 00:01:38.400 --> 00:01:39.997 browser, being able to kind of click 00:01:40.080 --> 00:01:45.517 through applications that are otherwise hard to to have an AI operate. 00:01:45.600 --> 00:01:48.837 And so, I think that we're really seeing the emergence of this new way of using a 00:01:48.920 --> 00:01:52.757 computer, and it starts with this kind of intelligence at the core. 00:01:52.840 --> 00:01:54.357 >> When we spoke last, you mentioned that 00:01:54.440 --> 00:01:56.677 this was effectively the culmination of 00:01:56.760 --> 00:01:59.917 a two-year research process. So, was 00:02:00.000 --> 00:02:04.437 this planned two years ago? Is that how far back OpenAI plans? 00:02:04.520 --> 00:02:06.477 >> I would say that yes, we do have very 00:02:06.560 --> 00:02:08.997 long horizons for how we plan. Now, one 00:02:09.080 --> 00:02:11.917 note is that we stack together many 00:02:12.000 --> 00:02:17.317 research ideas and bets on a variety of timescales. And so, the way to think 00:02:17.400 --> 00:02:20.637 about it is that we are making constant progress across every single part of the 00:02:20.720 --> 00:02:24.117 stack. And so, what GPT-5.5 represents 00:02:24.200 --> 00:02:27.877 is not an end point. In many ways, it's a beginning point. It's really a step 00:02:27.960 --> 00:02:32.477 towards the kinds of models that we see coming over even just upcoming months. 00:02:32.560 --> 00:02:36.597 And I think that you should expect that we are going to have even larger 00:02:36.680 --> 00:02:38.757 improvements in the capability across a 00:02:38.840 --> 00:02:43.797 wide variety of of of these uh uh of these uh aspects of what the model 00:02:43.880 --> 00:02:47.317 can do. And that's something that I think will be very exciting, and we're 00:02:47.400 --> 00:02:50.797 just always thinking about how can we make what we're producing more useful 00:02:50.880 --> 00:02:55.077 for real-world use, for real users, and real applications. 00:02:55.160 --> 00:02:58.757 >> Can you share specifically what those uh aspects are that we should be looking 00:02:58.840 --> 00:03:01.357 out for over the next few months? 00:03:01.440 --> 00:03:04.237 Um if this is the beginning, what is it the beginning of? 00:03:04.320 --> 00:03:08.517 >> Well, I think that the big vision we have, and you can see it reflected in 00:03:08.600 --> 00:03:11.837 many things, not just the models, but the the kind of you know, you think 00:03:11.920 --> 00:03:16.317 about the models as the brain, you can think about the systems and the 00:03:16.400 --> 00:03:19.397 harnesses like Codex and the applications like the super app as 00:03:19.480 --> 00:03:23.357 almost the body around it to make it into a useful AI. And that's really 00:03:23.440 --> 00:03:26.957 what's happening is a shift from language models being the thing that is 00:03:27.040 --> 00:03:32.077 produced by labs like ourselves to an AI that's actually useful. It's actually an 00:03:32.160 --> 00:03:33.437 assistant that's out there trying to 00:03:33.520 --> 00:03:36.357 solve your goal, that's really operating 00:03:36.440 --> 00:03:40.117 according to your instruction. And you 00:03:40.200 --> 00:03:44.277 can see right now Codex is becoming 00:03:44.360 --> 00:03:45.997 this app that's not just for the coders. 00:03:46.080 --> 00:03:48.517 It's really for anyone using a computer. 00:03:48.600 --> 00:03:52.957 And that it's not perfect, right? That there are still some tasks where that it 00:03:53.040 --> 00:03:56.037 should be able to do it and it doesn't quite get it right. Sometimes the 00:03:56.120 --> 00:03:59.277 personality isn't quite what you wanted, right? That it doesn't quite, you know, 00:03:59.360 --> 00:04:02.597 it's like extremely powerful and do out there doing a lot of really amazing 00:04:02.680 --> 00:04:06.277 things, but the way it communicates back to you that you have to still spend some 00:04:06.360 --> 00:04:08.757 time really trying to read through, okay, exactly how did it solve this 00:04:08.840 --> 00:04:12.957 problem? And so, these aspects, we know exactly how to make them much better. 00:04:13.040 --> 00:04:16.197 And we've already had a pretty remarkable improvement from 5.4 00:04:16.280 --> 00:04:20.357 to 5.5. I think we're going to have even more remarkable improvements across 00:04:20.440 --> 00:04:24.397 every single aspect of what makes these models useful. And one thing to know 00:04:24.480 --> 00:04:28.917 internally is that we think a lot about the end application. Like that is one 00:04:29.000 --> 00:04:31.117 thing that changed for us over the past 00:04:31.200 --> 00:04:34.157 12, you know, 18 months, something like 00:04:34.240 --> 00:04:38.277 that, is that we used to really just be focused on let's be let's improve on the 00:04:38.360 --> 00:04:43.037 benchmarks, let's make these models more cerebrally capable, but we now are 00:04:43.120 --> 00:04:45.757 really focused on let's bring them to real-world application. Let's think 00:04:45.840 --> 00:04:50.357 about finance, sales, marketing, every single function that someone uses a 00:04:50.440 --> 00:04:53.597 computer, how can we help with their computer work? How can we actually make 00:04:53.680 --> 00:04:56.157 the model have not just the theoretical 00:04:56.240 --> 00:05:00.637 capability to help, but is actually experienced those kinds of tasks, it's 00:05:00.720 --> 00:05:02.197 actually been able to see what good 00:05:02.280 --> 00:05:04.837 looks like. And I think that the place 00:05:04.920 --> 00:05:08.117 we're going is one where you as a person 00:05:08.200 --> 00:05:12.597 doing work that you are the overseer, you are the the CEO of almost this 00:05:12.680 --> 00:05:16.237 autonomous corporation or you know, of this this fleet of agents perhaps is 00:05:16.320 --> 00:05:20.237 more is is the way to say it. And that they are operating according to 00:05:20.320 --> 00:05:22.837 your goals. Now, you are still accountable, right? You're still in the 00:05:22.920 --> 00:05:26.677 driver's seat. You're still the person who thinks about, well, is this what I 00:05:26.760 --> 00:05:30.877 actually wanted? Was this work up to up to standard? But that the details of 00:05:30.960 --> 00:05:34.557 exactly what buttons were clicked and exactly the kind of code that was 00:05:34.640 --> 00:05:37.837 written or exactly how the formula in the spreadsheet works, that you can 00:05:37.920 --> 00:05:42.357 abstract yourself from those if they're not important to the evaluation of 00:05:42.440 --> 00:05:45.397 whether or not something was what you wanted. And so, I think it's like 00:05:45.480 --> 00:05:48.477 increasing leverage for every worker. 00:05:48.560 --> 00:05:51.717 >> Okay, uh let me take my best guess as to what's happening and you tell me how 00:05:51.800 --> 00:05:55.237 close I am. I mean, I'm thinking about this. This is like a like you mentioned 00:05:55.320 --> 00:05:57.877 a combination of two years of work. 00:05:57.960 --> 00:06:00.797 There's two different types of I mean not to tell you you know this but for 00:06:00.880 --> 00:06:04.677 our audience two different types of AI training. There's the pre-training or at 00:06:04.760 --> 00:06:07.917 least the ones that have been pertinent for these models. The pre-training where 00:06:08.000 --> 00:06:11.117 you just make the model generally smart by having it predict the next word and 00:06:11.200 --> 00:06:15.477 then reinforcement training where you have it like go out and actually take 00:06:15.560 --> 00:06:18.477 you know try to accomplish different tasks and you reward it when it does a 00:06:18.560 --> 00:06:21.517 good job with those tasks and effectively it sort of 00:06:21.600 --> 00:06:25.077 teaches it or it learns how to how to do those tasks. Um 00:06:25.160 --> 00:06:28.997 is is what you're saying basically that like this is the first result that we're 00:06:29.080 --> 00:06:33.397 seeing where open AI has just loaded a ton of reinforcement learning on task 00:06:33.480 --> 00:06:37.357 specific stuff uh into this model and that's what's producing the results 00:06:37.440 --> 00:06:40.757 you're talking about. >> Well, I would actually say it a little 00:06:40.840 --> 00:06:44.597 differently. I would say that there's many steps in the pipeline, right? That 00:06:44.680 --> 00:06:47.477 there's pre-training, mid-training, reinforcement learning. There's you 00:06:47.560 --> 00:06:50.637 know, the data collection. There's like a lot of these different things that all 00:06:50.720 --> 00:06:54.157 come together to produce the end result and the way in which it's connected to 00:06:54.240 --> 00:06:58.277 the world, that's also very key to making it useful. And the thing that I'm 00:06:58.360 --> 00:07:00.197 really saying is we have been investing 00:07:00.280 --> 00:07:03.877 on every single one of these and have a 00:07:03.960 --> 00:07:06.037 repeatable we have like a team, right? 00:07:06.120 --> 00:07:09.277 That it's not just about individuals working on these pieces but a team that 00:07:09.360 --> 00:07:12.357 really comes together and looks across the whole stack to say, "How do we make 00:07:12.440 --> 00:07:15.757 this more useful for real world applications?" And so it's not really 00:07:15.840 --> 00:07:20.917 any one thing that we do. It's really about the the overall effort of trying 00:07:21.000 --> 00:07:23.917 to like if you think about if you're building a car, right? That there's it's 00:07:24.000 --> 00:07:26.197 not just about do you have like a better engine, right? You could build a great 00:07:26.280 --> 00:07:30.197 engine but if the rest of the car is not up to the quality level of the engine, 00:07:30.280 --> 00:07:33.517 it's not going to matter. And so I think that that is the real innovation. It's 00:07:33.600 --> 00:07:37.637 really the end-to-end co-design and all coming together in a repeatable fashion 00:07:37.720 --> 00:07:41.397 to make these models better and better for our users. 00:07:41.480 --> 00:07:45.157 >> Um you were on a media call earlier today with myself and a number of 00:07:45.240 --> 00:07:49.117 members of the press and uh one of the interesting things that you said or 00:07:49.200 --> 00:07:51.797 basically I think you said this right off the bat is that the model more 00:07:51.880 --> 00:07:54.957 intuitively knows uh what you want and you don't have to 00:07:55.040 --> 00:07:59.357 spell it out exactly um as you as you would in the past. Uh here's a tweet 00:07:59.440 --> 00:08:01.557 from from Rune, "There are early signs 00:08:01.640 --> 00:08:06.397 of 5.5 being a competent AI research partner. Several researchers let 5.5 run 00:08:06.480 --> 00:08:10.677 variations of experiments overnight given only a high-level uh algorithmic 00:08:10.760 --> 00:08:14.797 idea waking up to find a completed sweep dashboards and samples never having 00:08:14.880 --> 00:08:17.677 touched the code or terminal at all." 00:08:17.760 --> 00:08:22.117 Um just uh if you can answer briefly on this, a two-parter. How do you do that? 00:08:22.200 --> 00:08:26.317 Uh and um does that mean prompt engineering is dead? 00:08:26.400 --> 00:08:29.877 >> Um number one, I think it really comes down to when we 00:08:29.960 --> 00:08:33.797 say there's a new class of capability, new class of intelligence, that's really 00:08:33.880 --> 00:08:35.077 what we mean, right? The models are 00:08:35.160 --> 00:08:39.237 becoming much more intuitive to use 00:08:39.320 --> 00:08:43.357 because they have deeper understanding of what it is you're asking of them, 00:08:43.440 --> 00:08:46.517 right? That they really look at the context, try to understand and puzzle 00:08:46.600 --> 00:08:50.477 out, "What am I being asked to do?" And it really makes you realize, you know, 00:08:50.560 --> 00:08:52.877 to the second part, is prompt engineering dead, which I actually think 00:08:52.960 --> 00:08:57.997 that prompt engineering in some ways may be even more vibrant than before. Um 00:08:58.080 --> 00:09:01.877 but you spend so much time right now trying to explain to your computer what 00:09:01.960 --> 00:09:05.277 you even want. You try to like pack in this context and be like, "Well, here's 00:09:05.360 --> 00:09:08.157 what's going on, here's the situation, here's the thing I want from you." And 00:09:08.240 --> 00:09:10.917 you're like, "Why do I have to explain this to my computer? Right? Like the 00:09:11.000 --> 00:09:13.997 whole thing is the computer should be doing the work to help me. Like I don't 00:09:14.080 --> 00:09:16.957 want to have to be sort of, you know, breaking down the task, trying to 00:09:17.040 --> 00:09:19.957 explain to it step by step how to do things. I want to point it in a 00:09:20.040 --> 00:09:23.917 direction and I want it to be able to take care of the details and to get me 00:09:24.000 --> 00:09:27.357 the result. Again, in a way that I can observe and kind of provide feedback 00:09:27.440 --> 00:09:31.517 along the way, but I want it to be the driver of the of the of those like 00:09:31.600 --> 00:09:35.997 low-level execution. And so I think that in some ways where prompt engineering is 00:09:36.080 --> 00:09:39.837 going to go is it's going to be about you can get so much more out of these 00:09:39.920 --> 00:09:44.637 models with so much less effort, but with the same amount of effort 00:09:44.720 --> 00:09:47.797 you still have a multiplier. Think about how much more you could even get. And I 00:09:47.880 --> 00:09:50.877 think that we're just at the leading edge right now of seeing the ceiling of 00:09:50.960 --> 00:09:54.317 what is capable what even today's models are capable of. 00:09:54.400 --> 00:09:58.237 >> Okay. Let me briefly speak with you about the economics of building a model 00:09:58.320 --> 00:10:02.797 like this. There's been this pattern where these big massive models now 00:10:02.880 --> 00:10:06.517 you're not saying how much money your computer used used you've used to train 00:10:06.600 --> 00:10:09.397 this but I think we can be safe in assuming 00:10:09.480 --> 00:10:12.477 it was a lot and there's been this pattern where these massive models come 00:10:12.560 --> 00:10:16.237 out they get distilled by open source 00:10:16.320 --> 00:10:19.997 model makers and then open source is 00:10:20.080 --> 00:10:23.837 just a couple months behind the leading foundational models 00:10:23.920 --> 00:10:27.717 um and you know I guess like when the investment was smaller being a couple 00:10:27.800 --> 00:10:31.237 months ahead you know mattered a lot but I'm curious 00:10:31.320 --> 00:10:35.677 now that the investment is so big and the models are the capabilities are 00:10:35.760 --> 00:10:39.357 increasing you know fairly dramatically you know as 00:10:39.440 --> 00:10:43.397 you go how is this defensible in the long term 00:10:43.480 --> 00:10:46.597 if you're just going to have that pattern repeat over and over? 00:10:46.680 --> 00:10:48.077 >> Well I look at it a little differently. 00:10:48.160 --> 00:10:52.557 Like I think that the real investment that we are making is into 00:10:52.640 --> 00:10:57.157 that end-to-end code design right of having a system a system of people right 00:10:57.240 --> 00:11:01.557 who are producing this technology right a way of working together and some of 00:11:01.640 --> 00:11:03.157 this is about how you leverage these 00:11:03.240 --> 00:11:05.317 massive supercomputers produce these 00:11:05.400 --> 00:11:10.637 models. Now it is also the case that it's not as simple as you can take the 00:11:10.720 --> 00:11:13.637 output of these models and distill them and you have exactly the model the same 00:11:13.720 --> 00:11:16.917 capability. It's just smaller and could run fast. If 00:11:17.000 --> 00:11:19.797 that were the case we would just do that and then we would also have a model that 00:11:19.880 --> 00:11:23.557 would be you know much more easy to serve it in many ways and of 00:11:23.640 --> 00:11:26.717 course there's a lot of art behind distillation there's a lot of great 00:11:26.800 --> 00:11:30.477 things there but the point that I'm getting at is that the real thing that 00:11:30.560 --> 00:11:34.637 we are investing in is the machine that makes the machine. Now, the at the 00:11:34.720 --> 00:11:37.757 deployment side, we think a lot about safeguards. We think a lot about 00:11:37.840 --> 00:11:39.677 mitigations. And we do that for many, 00:11:39.760 --> 00:11:42.157 many different aspects of how these 00:11:42.240 --> 00:11:47.557 models could be misused um in real situations. And that's something 00:11:47.640 --> 00:11:50.917 that we have been investing in for many years. And we think about that across 00:11:51.000 --> 00:11:52.677 areas like cyber or thinking about that 00:11:52.760 --> 00:11:56.157 in areas like bio that we have a I a 00:11:56.240 --> 00:11:59.757 long-standing effort that you can see in our preparedness framework, which is 00:11:59.840 --> 00:12:03.837 public, about how we approach these kinds of uses of the model and how we 00:12:03.920 --> 00:12:07.597 try to make maximize the benefits, mitigate the risks. And so, I think it's 00:12:07.680 --> 00:12:10.077 a real motion that every piece of what 00:12:10.160 --> 00:12:14.637 we do needs to connect to the question of how do we continue to make progress, 00:12:14.720 --> 00:12:17.557 but also how do we make these models broadly available? Because that's 00:12:17.640 --> 00:12:20.637 something that we really believe in, that we believe this technology empowers 00:12:20.720 --> 00:12:24.837 people, and that we want it to benefit people and lift everyone up. 00:12:24.920 --> 00:12:27.397 >> Yeah. But, just to go back on that, um 00:12:27.480 --> 00:12:29.237 the pricing on this model is, I think, 00:12:29.320 --> 00:12:33.717 double uh last model, GPT-5.4. 00:12:33.800 --> 00:12:37.477 Um and so, from an economics or a business standpoint, the question would 00:12:37.560 --> 00:12:41.197 be, you know, let's say you keep on progressing, but because there's been 00:12:41.280 --> 00:12:43.997 all this infrastructure that infrastructure that's been put towards 00:12:44.080 --> 00:12:46.837 training the models, if open source uh 00:12:46.920 --> 00:12:48.837 can deliver not as good performance, but 00:12:48.920 --> 00:12:52.557 almost as good, uh and do it cheaper, um 00:12:52.640 --> 00:12:54.437 how do you handle that threat? 00:12:54.520 --> 00:12:57.877 >> Well, again, I I look at it a little differently. So, first of all, if you 00:12:57.960 --> 00:13:01.637 look at our history, which really is not driven by anything 00:13:01.720 --> 00:13:03.277 in competition. It's just like our our 00:13:03.360 --> 00:13:07.077 own sort of progress and and desire, we 00:13:07.160 --> 00:13:11.677 have dropped prices on the same level of intelligence year over year, sometimes 00:13:11.760 --> 00:13:13.997 by literally a factor of 100. Right? 00:13:14.080 --> 00:13:17.237 It's like at least an order of magnitude year over year, sometimes literally 100. 00:13:17.320 --> 00:13:19.757 But, the thing that keeps happening, it's real Jevons paradox, where it's 00:13:19.840 --> 00:13:22.357 like you lower the cost of something, 00:13:22.440 --> 00:13:25.077 way more activity happens, right? And I 00:13:25.160 --> 00:13:29.797 think that what we keep seeing is that there are returns to intelligence, 00:13:29.880 --> 00:13:33.797 right? That for the kinds of tasks that these models are now capable of doing, 00:13:33.880 --> 00:13:37.277 that a little bit more intelligence goes a long way. And I think that is the 00:13:37.360 --> 00:13:41.917 story of 5-5, that in some ways you can almost look at it as like, oh, there's 00:13:42.000 --> 00:13:44.597 just an incremental improvement in intelligence, but I think there's going 00:13:44.680 --> 00:13:48.397 to be a massive improvement in terms of what people use it for. And by the way, 00:13:48.480 --> 00:13:51.997 I actually think that incremental is actually very much an understatement for 00:13:52.080 --> 00:13:56.877 this model relative to 5-5. You know, it's a 0.1 improvement in some ways, um 00:13:56.960 --> 00:14:01.197 but I think that that that actually really undersells the magic that we see 00:14:01.280 --> 00:14:04.477 within this model and that the that our early testers have have really seen in 00:14:04.560 --> 00:14:08.877 their practical work. >> So, if people see these numbers and they say, 00:14:08.960 --> 00:14:09.597 um 00:14:09.680 --> 00:14:13.917 uh there's IPO pressure on uh on OpenAI, 00:14:14.000 --> 00:14:16.277 and therefore the you know, we've been 00:14:16.360 --> 00:14:21.077 getting a great deal on intelligence and the free ride is over. 00:14:21.160 --> 00:14:24.437 You would argue against that. >> I I Yeah, look, the way I think about 00:14:24.520 --> 00:14:25.957 this is that we have a very simple 00:14:26.040 --> 00:14:28.637 business in some ways, right? We rent, 00:14:28.720 --> 00:14:31.717 build, buy compute, and we resell it 00:14:31.800 --> 00:14:35.877 with some positive margin. And as long as it's 00:14:35.960 --> 00:14:39.517 you know, positive operating margin, and as long as there's scalable demand 00:14:39.600 --> 00:14:42.757 for intelligence, which I think is true as long as there's problems to solve, 00:14:42.840 --> 00:14:45.197 like no one's going to run out of problems to solve, and we've seen this 00:14:45.280 --> 00:14:48.597 at every step that the demand outstrips our supply, then we can scale that 00:14:48.680 --> 00:14:52.837 compute all day. And I think that that in my mind, that's the main directive 00:14:52.920 --> 00:14:56.837 that I that I ask of the team. It's just like, just think about we need to add 00:14:56.920 --> 00:14:58.557 value on top of the raw compute and make 00:14:58.640 --> 00:15:00.237 sure that we are at positive operating 00:15:00.320 --> 00:15:03.957 margin on it, and that that is something 00:15:04.040 --> 00:15:08.517 where it's actually not even about like like the different competition in the 00:15:08.600 --> 00:15:11.797 marketplace. It's just a question of can you 00:15:11.880 --> 00:15:16.917 like have compute that gets turned into intelligence, and that's just how, you 00:15:17.000 --> 00:15:20.157 know, that it does that at at a you know, slightly improved uh you know, 00:15:20.240 --> 00:15:23.877 value coming out relative to the cost going in. And I think that that is 00:15:23.960 --> 00:15:26.597 something where again we're always trying to make more 00:15:26.680 --> 00:15:30.277 efficient models, but then we just want more of them, and then we want the more 00:15:30.360 --> 00:15:33.237 intelligent models, and regardless of where they're coming from, it's kind of 00:15:33.320 --> 00:15:35.717 all the same compute that's going in. 00:15:35.800 --> 00:15:39.037 And so, I think that it's actually a great like competition this marketplace 00:15:39.120 --> 00:15:41.597 has been great for innovation, um but I 00:15:41.680 --> 00:15:46.277 think that it's actually something where it's driving more usage and more overall 00:15:46.360 --> 00:15:51.197 spend in the ecosystem, and you can see that in the revenue numbers of us and uh 00:15:51.280 --> 00:15:53.037 you know, others in this industry. 00:15:53.120 --> 00:15:55.837 >> Okay, I want to take a quick break and come back and talk to talk with you 00:15:55.920 --> 00:16:00.797 about cybersecurity, trust, and whatever else we can get to in our time in this 00:16:00.880 --> 00:16:04.637 emergency show. We'll be back right after this. And we're back here on Big 00:16:04.720 --> 00:16:07.157 Technology Podcast with OpenAI president 00:16:07.240 --> 00:16:09.277 and co-founder Greg Brockman. 00:16:09.360 --> 00:16:12.557 Um Greg, let me ask you about uh the 00:16:12.640 --> 00:16:15.517 cybersecurity implications here. Um two 00:16:15.600 --> 00:16:18.197 very different approaches between OpenAI 00:16:18.280 --> 00:16:20.597 and Anthropic. Anthropic's 00:16:20.680 --> 00:16:23.397 uh latest massive model, Mythos, uh is 00:16:23.480 --> 00:16:25.757 not released to the public. Um 00:16:25.840 --> 00:16:30.437 this one, you know, Spud or 5.5, is released to the public. 00:16:30.520 --> 00:16:33.637 I mean, let me just ask you straight up, is there a chance that uh releasing this 00:16:33.720 --> 00:16:37.957 powerful model into you know, into the public without this like step-by-step uh 00:16:38.040 --> 00:16:40.997 practice could lead to some major cyber attacks? 00:16:41.080 --> 00:16:42.917 >> Well, I I actually have a different view 00:16:43.000 --> 00:16:46.677 on the premise of the question. So, the thing to understand is that we have been 00:16:46.760 --> 00:16:48.477 investing in cyber safeguards and 00:16:48.560 --> 00:16:51.677 cybersecurity as a part of our 00:16:51.760 --> 00:16:53.877 preparedness framework for years, right? 00:16:53.960 --> 00:16:57.597 The This is something we have invested in far ahead of having the kinds of 00:16:57.680 --> 00:17:01.517 capabilities we see coming. And so, we have been taking a very deliberate 00:17:01.600 --> 00:17:05.677 step-by-step approach. You can see even just over the past couple weeks where 00:17:05.760 --> 00:17:09.637 we've expanded our trusted access for cyber program, and in general, we 00:17:09.720 --> 00:17:11.517 believe in ecosystem resilience, right? 00:17:11.600 --> 00:17:14.197 That we think that you do want to go step-by-step, that these models are 00:17:14.280 --> 00:17:18.037 going to continuously better. We have line of sight to even more capable ones, 00:17:18.120 --> 00:17:22.157 and that you want to be able to I put these models in the 00:17:22.240 --> 00:17:24.957 hands of defenders to make sure that you're able to protect critical 00:17:25.040 --> 00:17:29.237 infrastructure. And we believe in in that resilience of as you can bring 00:17:29.320 --> 00:17:32.637 these models into people's hands that that then they're able to explore in 00:17:32.720 --> 00:17:36.957 ways that you would not be able to without that kind of access. And so, you 00:17:37.040 --> 00:17:40.677 kind of want this graduated approach and to make sure that you are moving down 00:17:40.760 --> 00:17:44.637 that pipeline as you can bring in additional safeguards uh in order to 00:17:44.720 --> 00:17:47.997 make sure that you can maximize the benefits and mitigate the risks. And so, 00:17:48.080 --> 00:17:50.637 we've really taken a deliberate approach. I think our team has been 00:17:50.720 --> 00:17:54.917 working incredibly hard to think through the cyber implications of this model. 00:17:55.000 --> 00:17:58.797 Um we also believe in iterative deployment. Uh that's part of this 00:17:58.880 --> 00:18:01.757 really bringing the models as they continuously get better. And we believe 00:18:01.840 --> 00:18:04.997 in democratic access. We believe that ultimately the goal of 00:18:05.080 --> 00:18:08.597 creating this technology is to empower people to ensure that it does benefit 00:18:08.680 --> 00:18:12.437 all of humanity. And so, we are constantly trying to solve for how do we 00:18:12.520 --> 00:18:16.677 safely and responsibly bring this technology to bear in the world in a 00:18:16.760 --> 00:18:21.157 broad way. >> Right. And um I think suffice it to say 00:18:21.240 --> 00:18:24.757 that um your team hasn't been fans of the way that Anthropic's deployed 00:18:24.840 --> 00:18:28.237 mythos. This is a quote from Sam. Uh it's clearly incredible marketing to say 00:18:28.320 --> 00:18:31.357 we have built a bomb, we're about to drop it on your head, we will sell you a 00:18:31.440 --> 00:18:34.677 bomb shelter for a hundred million to run all your to run across all your 00:18:34.760 --> 00:18:38.517 stuff, but only if we pick you as a customer. Um 00:18:38.600 --> 00:18:42.397 let me talk through the other case and then get your response. 00:18:42.480 --> 00:18:46.677 Um the other case would be there are you can't account for everything, and there 00:18:46.760 --> 00:18:50.197 are clearly going to be some vulnerabilities that can't that will 00:18:50.280 --> 00:18:53.997 only be found by people or entities deploying this and 00:18:54.080 --> 00:18:57.797 looking for them. So, maybe it makes sense to start with a trusted group of 00:18:57.880 --> 00:19:00.557 testers before you deploy it before you 00:19:00.640 --> 00:19:03.037 deploy it broadly. What do you think? 00:19:03.120 --> 00:19:07.517 >> Well, I believe the correct answer here is subtle, and I think it is rooted in 00:19:07.600 --> 00:19:11.877 the technical specifics of what you have in front of you and many, many factors, 00:19:11.960 --> 00:19:15.437 right? You need to think about how are the models progressing, right? Not just 00:19:15.520 --> 00:19:19.397 your own capabilities, but others in the ecosystem. You need to think about 00:19:19.480 --> 00:19:22.877 what kind of benefit do you get from having a small group that has access and 00:19:22.960 --> 00:19:26.917 are able to have, you know, are they able to have high leverage by by being 00:19:27.000 --> 00:19:30.237 able to find and produce patches, but then how do you actually coordinate the 00:19:30.320 --> 00:19:32.597 disclosure of those across an industry? 00:19:32.680 --> 00:19:35.797 And so there's a lot of factors that go into it. I think that the true answer is 00:19:35.880 --> 00:19:39.997 like if either extreme is not quite right. There are tools that can be 00:19:40.080 --> 00:19:43.197 applied to a specific situation. And I think that this is not the first time 00:19:43.280 --> 00:19:44.437 we've had to think about this problem. 00:19:44.520 --> 00:19:47.597 It's not the last time we will have to think about it. But one thing to note is 00:19:47.680 --> 00:19:50.237 that we have had our model in the hands 00:19:50.320 --> 00:19:55.317 of defenders um for some time that we've been building up our trusted access 00:19:55.400 --> 00:19:58.317 program. That the model that we're releasing is actually not 00:19:58.400 --> 00:20:01.437 cyber-permissive, right? That it actually has a number of safeguards 00:20:01.520 --> 00:20:03.717 built into it. And that you can then 00:20:03.800 --> 00:20:08.477 have a gap between what you're privately sharing, testing, those kinds of things. 00:20:08.560 --> 00:20:11.517 And so I think my my short answer is like it's there's definitely these 00:20:11.600 --> 00:20:16.277 different schools of thought in terms of values of is the value that you want to 00:20:16.360 --> 00:20:20.237 get these models into people's hands and empower them or is the value that you 00:20:20.320 --> 00:20:23.397 kind of want them to be centralized and controlled and that you don't want them 00:20:23.480 --> 00:20:27.437 in people's hands? That is something that is a maybe underlying tension in 00:20:27.520 --> 00:20:28.997 some of these debates. But I think that 00:20:29.080 --> 00:20:34.477 the tactics, right, the you know, that those almost flow from the details and 00:20:34.560 --> 00:20:38.517 that they can be informed by these values. But either extreme, reflexively, 00:20:38.600 --> 00:20:42.237 I don't think will yield the best outcome for the world. 00:20:42.320 --> 00:20:46.197 >> Okay, I want to ask you about agents. Uh back to agents if we could. Um these 00:20:46.280 --> 00:20:50.317 agents work work the best if you sort of let them uh 00:20:50.400 --> 00:20:54.917 have a high degree of autonomy. I mean, sort of makes sense. Um 00:20:55.000 --> 00:20:58.477 so I'm just kind of curious to hear your perspective. As we get more agents that 00:20:58.560 --> 00:21:01.997 can do more things and access more files and work across programs, what is the 00:21:02.080 --> 00:21:05.437 proper amount of trust to put into agents right now? 00:21:05.520 --> 00:21:08.117 >> So, I think that right now, actually, 00:21:08.200 --> 00:21:11.957 agents tend to be quite reliable. Um and 00:21:12.040 --> 00:21:15.597 even things like prompt injections, I think that there's still holes there, 00:21:15.680 --> 00:21:18.037 but that we're patching them and that the models are becoming much more 00:21:18.120 --> 00:21:22.837 resilient. But, I also think that the flip side is that as these models be are 00:21:22.920 --> 00:21:27.077 given increasing responsibility and access to more important context, that 00:21:27.160 --> 00:21:31.197 you need to have some answer for just like if you have employees, you know, if 00:21:31.280 --> 00:21:34.717 you have a team of five employees, they're all kind of trustworthy, 00:21:34.800 --> 00:21:39.317 fine. But, if you have 500,000 of the same employees, that some somehow those 00:21:39.400 --> 00:21:42.997 numbers, right, just like that there's the law of large numbers, that you start 00:21:43.080 --> 00:21:46.077 to worry about, "Okay, how do I have good governance and oversight?" Right. 00:21:46.160 --> 00:21:49.237 And so, this is something where as we're investing in these capabilities and 00:21:49.320 --> 00:21:52.837 making the super app more accessible not just to coders, but to to any person 00:21:52.920 --> 00:21:56.037 doing work with a computer, also investing in governance and oversight. 00:21:56.120 --> 00:21:59.797 And you can see this very concretely in Workspace agents, which we released 00:21:59.880 --> 00:22:04.077 recently. So, that's within your enterprise, you can now define agents. 00:22:04.160 --> 00:22:07.357 So, you get a hosted Codex harness in the cloud. You can hook up tools, you 00:22:07.440 --> 00:22:10.917 can hook up to your Slack, and it's doing work. It's like awesome. A lot of 00:22:11.000 --> 00:22:14.717 people use it. It's been very cool to see how sort of viral it goes within an 00:22:14.800 --> 00:22:17.077 organization when you see you use someone else's agent, you're like, 00:22:17.160 --> 00:22:18.917 "Wait, I can build one of these, too." 00:22:19.000 --> 00:22:21.797 And you can just fork it and and do your own thing. 00:22:21.880 --> 00:22:25.677 And then that's an opportunity to have great governance in that you can see 00:22:25.760 --> 00:22:29.477 that's baked into the product where your IT organization can see all the agents 00:22:29.560 --> 00:22:31.957 that have been created that for an agent, you can see the conversations 00:22:32.040 --> 00:22:35.477 it's had, and that you can think about exactly what the guardrails are around 00:22:35.560 --> 00:22:39.637 it. So, I think that the short answer is like you want to ramp the 00:22:39.720 --> 00:22:42.877 responsibility entrusted with the agent and the diversity of things that agents 00:22:42.960 --> 00:22:45.917 are doing together with security, 00:22:46.000 --> 00:22:50.677 safety, observability, oversight. And if you're not doing those in hand-in-hand, 00:22:50.760 --> 00:22:53.717 then I think that that that that's a little bit out of balance. And I think 00:22:53.800 --> 00:22:56.317 it's important to to think about both sides. 00:22:56.400 --> 00:22:59.797 >> Yeah, basically, go ahead, but be careful. 00:22:59.880 --> 00:23:03.357 >> But, you and really lean in, right? I think it's like as you scale, like you 00:23:03.440 --> 00:23:07.237 can prototype and that that it's just the nature of scale that starts to bring 00:23:07.320 --> 00:23:11.317 in the do you still have the ability to to oversee what's going on? So, you need 00:23:11.400 --> 00:23:15.037 to to kind of make sure at each step do you feel like you're calibrated, you 00:23:15.120 --> 00:23:17.317 understand what your teams are up to. 00:23:17.400 --> 00:23:19.197 >> Greg, let's end with this. Um you called 00:23:19.280 --> 00:23:22.597 this a compute-powered economy. What does that mean? 00:23:22.680 --> 00:23:24.917 >> Well, I think we are heading to a world 00:23:25.000 --> 00:23:27.677 where the more compute is poured into a 00:23:27.760 --> 00:23:29.757 problem, the faster that problem will be 00:23:29.840 --> 00:23:34.757 solved. And that the ceiling of problem that can be solved depends on how much 00:23:34.840 --> 00:23:37.877 compute is available. And you think about things like drug discovery, right? 00:23:37.960 --> 00:23:39.837 Being able to solve complex diseases. 00:23:39.920 --> 00:23:43.717 Like, those are Solving complex diseases like Alzheimer's is kind of outside of 00:23:43.800 --> 00:23:47.757 humanity's reach right now. We've never really done it. But, 00:23:47.840 --> 00:23:49.037 imagine a world where you can take a 00:23:49.120 --> 00:23:51.557 gigawatt data center and have it just 00:23:51.640 --> 00:23:55.197 think about how to solve Alzheimer's. 00:23:55.280 --> 00:23:59.037 For a month, for a year, however long it takes. And it may not be literally just 00:23:59.120 --> 00:24:03.037 cerebrally solving this problem, but it may have to consult with world experts, 00:24:03.120 --> 00:24:06.277 maybe it has to suggest experiments to get run in a wet 00:24:06.360 --> 00:24:11.397 lab. But, if you can actually solve such a problem, that would be such a 00:24:11.480 --> 00:24:14.517 transformatively positive thing for humanity. And I think we're heading to a 00:24:14.600 --> 00:24:18.717 world where that is how important problems get solved. And that is how 00:24:18.800 --> 00:24:22.797 tasks in your daily life can also be solved. Whether it's having an agent 00:24:22.880 --> 00:24:26.877 that knows you, that has your personal context, that is trustworthy, that you 00:24:26.960 --> 00:24:30.037 can ask for advice on health and you get back 00:24:30.120 --> 00:24:33.917 trustworthy information. Um And that's just a thing that's a 00:24:34.000 --> 00:24:37.237 smartphone that's that's in your pocket, right? You can just talk to and it'll be 00:24:37.320 --> 00:24:40.317 out there doing things. It proactively knows what are your goals, what are your 00:24:40.400 --> 00:24:42.957 interest, and how it can help you. And I 00:24:43.040 --> 00:24:45.077 think that big and small compute is 00:24:45.160 --> 00:24:48.157 going to be the resource that shows how 00:24:48.240 --> 00:24:52.717 much computers can be used to help people, to do work on behalf of people. 00:24:52.800 --> 00:24:55.677 And I think we're heading to that world and it's one that we're all building 00:24:55.760 --> 00:24:58.797 collectively. >> Yeah, and that I think would explain the 00:24:58.880 --> 00:25:03.277 massive investments that you've led making these big infrastructure bets. 00:25:03.360 --> 00:25:06.517 >> Still not enough. We're going to feel the scarcity. We're going to feel it. 00:25:06.600 --> 00:25:09.197 We're feeling it already. You can sense it right now [music] on people who are 00:25:09.280 --> 00:25:12.157 trying to use these agents and just simply cannot, you know, hitting the 00:25:12.240 --> 00:25:15.317 rate limits. Um so we're working on behalf of our customers [music] on 00:25:15.400 --> 00:25:18.237 behalf of of everyone who wants to use these agents to ensure that there is 00:25:18.320 --> 00:25:21.637 enough and I don't think we're going to get there. We're going to do our best, 00:25:21.720 --> 00:25:24.997 but I think that we are headed to a world of compute scarcity and uh again, 00:25:25.080 --> 00:25:28.117 I think this is something where >> [music] >> we can all contribute to trying to help 00:25:28.200 --> 00:25:30.917 there just be more availability of this in the world. 00:25:31.000 --> 00:25:33.557 >> Greg, busy day. Always appreciate [music] your time. Always great to speak 00:25:33.640 --> 00:25:35.477 with you. Thanks again for coming on. 00:25:35.560 --> 00:25:38.600 >> Likewise. Great chatting.