Demis Hassabis: Agents, AGI & The Next Big Scientific Breakthrough

[00:00] continual learning, long-term reasoning,

[00:02] [music] uh some aspects of memory, these

[00:05] are still unsolved. I think all of these

[00:07] are going to be required for AGI.

[00:09] Depending on what your AGI timeline is,

[00:11] you know, mine's like 2030 or something

[00:13] like this, then [music] if you start off

[00:15] on a deep tech journey today, you have

[00:18] to just consider AGI appearing in the

[00:21] middle of that journey. It's not bad

[00:22] necessarily, but you have to take that

[00:24] into account. [music] You have to have

[00:26] an active system uh that can actively

[00:29] solve problems for you to get to AGI.

[00:31] So, agents are that path, and I think

[00:33] we're just getting going.

[00:39] Demis Hassabis has had one of the most

[00:43] unusual careers in tech. He was a chess

[00:46] prodigy as a kid, then designed his

[00:49] first hit video game, Theme Park, at 17.

[00:53] He then went back to school, got a PhD

[00:56] in cognitive neuroscience, published

[00:58] foundational work on how memory and

[00:59] imagination work in the brain, and then

[01:02] in 2010 co-founded DeepMind with one

[01:05] mission, solve intelligence.

[01:08] And I think they've done it.

[01:11] Since then,

[01:13] uh his lab has gone on to do things most

[01:15] people thought were decades away.

[01:17] AlphaGo beat a world champion at Go.

[01:19] AlphaFold cracked protein structure

[01:21] prediction, a 50-year grand challenge in

[01:24] biology, and they gave it away for free

[01:27] to every scientist on Earth.

[01:29] That work won him the Nobel Prize in

[01:31] chemistry last year. Today, Demis leads

[01:34] Google DeepMind, where he's building

[01:37] Gemini and pushing toward the same goal

[01:39] he set when he was a teenager,

[01:42] artificial general intelligence. Please

[01:44] welcome Demis Hassabis.

[01:53] So, you've been thinking about AGI

[01:54] longer than almost anyone. Uh when you

[01:56] look at the current paradigm,

[01:58] large-scale pre-training, RLHF, chain of

[02:00] thought, how much of the final

[02:02] architecture for AGI do you think we

[02:05] already have, and what's fundamentally

[02:07] missing right now? Well, first of all,

[02:09] thank thanks, Gary, for that great

[02:10] introduction, and it's great to be here.

[02:12] Thanks for for welcoming here. It's

[02:13] amazing space, actually. I'll have to

[02:15] come back here often. Very inspiring

[02:17] that you all get to work in in in this

[02:19] space. So, the question is I think the

[02:22] the components that you just mentioned,

[02:24] I'm pretty sure will be part of the

[02:26] final architecture for AGI. So, I think

[02:29] they've come such a long way now, uh and

[02:32] we've proven out so many things about

[02:33] what they can do.

[02:35] Uh I can't see a world in which we'll

[02:37] sort of realize in a couple of years

[02:39] this was a dead end. That doesn't make

[02:40] sense to me. But, there still might be

[02:42] one or two things missing on top of uh

[02:45] of of of what you've you know, what we

[02:47] already know works. So, um

[02:50] continual learning, long-term reasoning,

[02:52] uh some aspects of memory, these are

[02:55] still unsolved. Um and how to get the

[02:58] systems to be more consistent across the

[03:00] board. Um I think all of these are going

[03:02] to be required for AGI. Now, it might be

[03:05] that the existing techniques can just

[03:07] scale up to that with some innovation

[03:09] and some incremental innovation. Um but,

[03:12] it could be that there's still one or

[03:13] two big ideas left uh that need to be

[03:16] cracked. I don't think it's more than

[03:18] one or two if there are out there. And I

[03:20] think, you know, my betting is uh about

[03:23] 50/50 if that's the case. So, of course,

[03:25] at DeepMind at Google DeepMind we work

[03:27] on both those things. I guess that's all

[03:29] I mean, working with a bunch of

[03:31] identical systems. The wildest thing to

[03:32] me is to what degree It's the same

[03:35] weights ev- over and over. So, this idea

[03:37] of continual learning is so interesting

[03:39] because like you know, right now we're

[03:41] sort of cobbling it together with duct

[03:43] tape, you know? Yes. These dream cycles

[03:45] at night and things like that.

[03:47] >> Yeah. It's pretty cool, the dream

[03:48] cycles, and we we used to think about

[03:50] this with consolidation with episodic

[03:52] memory. It's actually that's what I

[03:53] studied for my PhD is how the

[03:54] hippocampus works and integrates, you

[03:57] know, new knowledge gracefully into the

[04:00] existing knowledge base. So, the brain

[04:02] does that amazingly well. It it it does

[04:04] it through you know, during sleep uh

[04:06] especially things like REM sleep,

[04:08] replaying back episodes that that are

[04:10] important so that you can learn from it.

[04:12] In fact, our very first Atari program

[04:15] DQN, one of the ways it was able to

[04:17] master Atari games was by doing

[04:19] experience replay. So, we sort of

[04:21] borrowed that from from neuroscience and

[04:23] replayed successful trajectories

[04:26] uh many times, you know, that's

[04:28] way back in 2013 now in the in the dark

[04:30] ages of AI. It was uh a really important

[04:33] thing. And and I agree with you, we're

[04:34] kind of using duct tape right now. So,

[04:36] like shove it all in the context window.

[04:38] Um this but this seems a bit

[04:40] unsatisfying, right? And actually, even

[04:42] though uh we're working on machines, not

[04:45] biological brains, and so you

[04:47] potentially you could have, you know,

[04:49] millions or tens of millions size

[04:51] context window or memory, and it can be

[04:53] perfect, there's still a cost to looking

[04:56] it up and finding the right thing uh

[04:59] that that's actually relevant for the

[05:01] specific uh decision you've got to make

[05:03] right now. And that's non-trivial that

[05:05] cost, even if you can potentially store

[05:07] it all. I think there's actually a lot

[05:09] of room for innovation in in areas like

[05:11] memory. Yeah. I mean, the one thing is

[05:13] like it feels like a million

[05:14] token context one is actually bigger

[05:17] than I mean, it's plenty big, honestly.

[05:18] You can do so.

[05:20] It's plenty big for for for most things

[05:22] that it should be used for. I mean, if

[05:24] you think about the context window is

[05:26] sort of equivalent to working memory,

[05:28] you know, humans have we have like a few

[05:30] digits, you know, it's like a a dozen

[05:32] digits maybe, you know, average of

[05:34] seven. We got million or, you know, 10

[05:36] million context windows, but the problem

[05:38] is is that we're trying to store

[05:40] everything in that. You know, things

[05:42] that aren't in not important, things

[05:43] that are wrong. It's pretty brute force

[05:45] currently, and that doesn't seem uh

[05:47] right. And then the problem is if you're

[05:49] an agent trying to try and process live

[05:51] video, and you're just going to naively

[05:53] record all the tokens, then actually a

[05:56] million tokens isn't that much. It's

[05:58] only like 20 minutes. So, actually you

[06:01] need more if you want something that's

[06:02] going to understand your, you know, your

[06:05] what's going on in your life over maybe

[06:06] a month or two. DeepMind has

[06:09] historically leaned into reinforcement

[06:11] learning and search.

[06:13] AlphaGo, AlphaZero, and MuZero. How much

[06:16] of that philosophy is actually embedded

[06:17] in how you're building Gemini today? Is

[06:21] RL still underrated? Yeah, I think

[06:24] potentially it is. It sort of goes in in

[06:26] abs and way waves. You know, we've

[06:28] worked on agents since the beginning of

[06:30] DeepMind. In fact, we also That's what

[06:32] we said we were working on. And so, all

[06:33] of the Atari work and AlphaGo, most

[06:36] specifically, they're agent systems. And

[06:38] what we meant by that is systems that

[06:40] are able to, you know, accomplish goals

[06:42] on their own.

[06:43] And make active decisions and and make

[06:46] plans. And so, of course, we were doing

[06:48] it in the domain of games to to to make

[06:51] it tractable. And then doing

[06:53] increasingly complex games, things like

[06:55] StarCraft after AlphaGo, AlphaStar. So,

[06:59] we basically did all the games that are

[07:01] out there.

[07:02] And then of course, the question is can

[07:04] you generalize those models to be world

[07:06] models or models of language, not just

[07:08] models of simple games or even complex

[07:11] games. And that's what the last few

[07:13] years has been about. But really, you

[07:15] can think of a lot of the things we're

[07:17] doing today, all the leading models with

[07:19] thinking modes and chain of thought

[07:21] reasoning as aspects of what was sort of

[07:23] pioneered with AlphaGo coming back now.

[07:26] And I actually think there's a lot of

[07:28] work we did back then that is relevant

[07:31] today. And we're sort of re-looking at

[07:33] some of those old ideas

[07:35] at scale today in a more general way.

[07:38] Including things like Monte Carlo tree

[07:39] search and other other ways of doing

[07:41] augmenting the RL

[07:43] on top of the the reinforcement learning

[07:45] we're ready to do today. And I think a

[07:47] lot of those ideas both from AlphaGo and

[07:49] AlphaZero are really really relevant to

[07:52] to where we are with today's foundation

[07:54] models. And I think a lot of that is

[07:56] what we're going to see of the advances

[07:58] the next few years. One question I would

[08:00] have like obviously today you need

[08:03] bigger and bigger models to be smarter

[08:05] and smarter, but then we're also seeing

[08:07] distillation working and then smaller

[08:09] models can be like quite a bit faster. I

[08:11] think you know, you guys have incredible

[08:13] flash models that are like nine like

[08:15] you're finding that they're 95% as good

[08:18] as the frontier and at like 1/10 the

[08:21] price. Is that right? I think that's one

[08:23] of our core strengths is I mean you have

[08:25] to build the biggest models to to to to

[08:27] have the frontier capabilities, but I

[08:30] think one of our biggest strengths has

[08:31] been

[08:32] distilling and packing that power into

[08:34] smaller and smaller models very quickly.

[08:37] Obviously we we you know, we invented

[08:39] the kind of distillation process and and

[08:41] people like Jeff and Oriol and and

[08:43] others and we're still world experts in

[08:46] that. And we also have a huge need to

[08:50] do it because we've got to serve the

[08:52] biggest probably AI surfaces there are.

[08:56] Obviously there's search with AI

[08:57] overviews and AI mode and there's Gemini

[08:59] app and now increasingly every single

[09:01] product at Google has you know, maps and

[09:04] YouTube and so on has some aspect of

[09:07] Gemini or Gemini related technology in

[09:09] it. And so that's billions of users a

[09:12] dozen more than a dozen billion user

[09:14] products

[09:15] and they have to be served extremely

[09:16] fast, extremely efficiently and cheaply

[09:18] and with low latency. So that that gives

[09:21] us a really important incentive to to

[09:24] make these flash and even smaller models

[09:26] flashlight models extremely efficient.

[09:29] And hopefully that ends up then being

[09:30] really useful for many of the workloads

[09:32] that all of you use for. I'm curious

[09:35] about how much smarter these smaller

[09:38] models can actually be. Like, are there

[09:39] limits to the distillation process?

[09:41] Like, could a 50B or 400B model be as

[09:45] smart as like a mythos for today? Yeah,

[09:48] I didn't I didn't see any I don't think

[09:49] we've got to any kind of or at least

[09:51] none of us know yet if we've got to any

[09:53] kind of informational limit. I mean,

[09:55] maybe at some point that will be the

[09:56] case where there's just an information

[09:58] density that can't we can't get beyond.

[10:00] But, I think for now there's the

[10:03] assumption we make is that you know, a

[10:05] year later one of our

[10:07] leading, you know, pro models or

[10:09] frontier models goes out, half a year

[10:12] later, a year later you'll have them in

[10:14] the the really tiny almost edge models.

[10:17] And you'll also see some of that

[10:18] goodness in our Gemma models, which

[10:20] hopefully you're all enjoying our Gemma

[10:21] four models, which I think are really

[10:23] amazing power for their sizes. So,

[10:26] again, that uses a lot of this these

[10:29] distillation techniques and and the idea

[10:31] of how to make things really efficient

[10:32] in these very small models. So, I don't

[10:34] really see any limit yet in terms of

[10:36] like some kind of theoretical limit. I

[10:37] think we're still pretty far off of

[10:39] that. That's a mean I mean that is

[10:40] really good.

[10:41] >> Yes.

[10:42] Uh you know, one of the weirder things

[10:43] that we're seeing right now is like

[10:45] engineers can do like 500 to 1,000 times

[10:48] the amount of work that they were doing

[10:50] like 6 months ago, I guess. I mean, the

[10:53] people in this room there are people who

[10:54] are doing about like a thousand X the

[10:56] work that like I Steve Yegge talks about

[10:59] this. It's like a thousand X the work

[11:01] that a Google engineer from the 2000s

[11:03] was doing. I think it's very exciting. I

[11:05] mean, I think models have many uses. One

[11:07] is obviously cost, but the speed can

[11:10] allow, you know, if you think about

[11:11] coding even or other things, you can

[11:13] iterate a lot faster. Also, especially

[11:15] if there's if you're collaborating with

[11:17] the system. I think there's a there's a

[11:20] a lot of need for having fast systems

[11:23] that maybe are not quite frontier. Like

[11:26] you said, like 95% 90%, but that's

[11:28] plenty good enough and actually gain

[11:29] back more than the 10% on the the

[11:32] iteration speed. So, and then the other

[11:34] big thing I think is running these

[11:36] things on the edge. Again, for

[11:37] efficiency reasons, but also for privacy

[11:40] and security reasons, too. Um if you

[11:42] think about different devices that you

[11:44] might run these systems on that that,

[11:47] you know, process very personal

[11:48] information, can also think about

[11:50] robotics, as well. Um you know, robots

[11:52] in your house. I think you're going to

[11:54] want very efficient, uh very powerful,

[11:57] uh local models, which maybe are

[11:59] orchestrated

[12:00] you know, with some bigger models,

[12:02] frontier models that are in the in the

[12:04] cloud, but you only delegate to that in

[12:06] certain circumstances. And perhaps you,

[12:09] you know, you process all of the

[12:11] audio-visual feed, let's say, locally,

[12:13] and that stays local. I could imagine uh

[12:16] that would be a very good sort of um end

[12:18] state. Y Combinator Startup School is

[12:20] back. We're hand-selecting the most

[12:22] promising builders in the world and

[12:24] flying them out to San Francisco on July

[12:26] 25th and 26th to discuss the cutting

[12:29] edge of tech. Apply now for a spot.

[12:31] Okay, back to the video. Going back to

[12:34] context and memory, models currently

[12:36] stateless, but, you know, continue like

[12:38] what would the developer experience even

[12:40] be like for someone who's using a

[12:42] continual learning model? Like, you

[12:44] know, any idea like how you'd steer it?

[12:46] I think it's really interesting. I think

[12:48] that's one of the not having continual

[12:50] learning currently is one of the things

[12:51] holding back agents from doing full

[12:55] uh tasks, you know? I think they're

[12:56] really useful for aspects of tasks right

[12:59] now, and you can patch them together and

[13:01] do some really cool things, but they

[13:03] don't adapt well with the context that

[13:06] you're in. And I think that's the

[13:08] missing piece for them being really kind

[13:10] of fire and forget, and they'll figure

[13:12] it out themselves. You know, I think

[13:14] they need to be able to learn um about

[13:16] the specific context um that you're

[13:19] going to put them in. So, um I think we

[13:22] have to crack that to get full general

[13:25] intelligence. Where are we on reasoning?

[13:27] So, models can do really impressive

[13:28] chain of thought now, but they still

[13:30] fail on things a smart undergrad

[13:32] wouldn't. What specifically needs to

[13:34] change and what progress do you expect

[13:36] in reasoning? There's a lot of

[13:39] innovation left in in think the thinking

[13:41] paradigms, I would say. Again, I think

[13:43] we're fairly we're doing fairly

[13:44] simplistic things, fairly brute force.

[13:48] One could imagine

[13:50] I think there's a lot of scope for

[13:51] example in monitoring the chain of

[13:52] thought, maybe interjecting midway

[13:55] through a thought process. I often get

[13:57] the impression with our systems and and

[13:59] our competitor systems that they're

[14:01] almost overthinking. They're almost

[14:03] getting into sort of loops of things.

[14:05] Like one thing I sometimes like to do is

[14:08] is play chess against Gemini. And you

[14:10] know, it's the all the leading

[14:12] foundation models are pretty poor at

[14:13] games, which is quite interesting. It's

[14:15] very

[14:17] cool to kind of look at the thinking

[14:18] traces cuz obviously these are can be a

[14:20] well-understood. You know, I can tell

[14:22] quite quickly if it's going off on a

[14:24] tangent and it's very sort of provable

[14:26] what the what the the thinking is doing,

[14:29] whether it's useful or not. And so, what

[14:32] we see is that, you know, sometimes it

[14:34] will it will it will consider a move. It

[14:36] will realize it's a blunder, but it

[14:38] can't find anything better, so it kind

[14:39] of goes back to that move and does it

[14:41] anyway. So, it you know, you just

[14:43] shouldn't be seeing that

[14:45] happening in a in a very precise

[14:47] reasoning system. So, there's just sort

[14:49] of huge gaps, I think, still, but it may

[14:52] only be one or two tweaks that are

[14:53] required to fix those kind of gaps just

[14:55] to be clear, but I think that's pretty

[14:57] pretty obvious they're all there. And

[14:59] that's why you get this kind of jagged

[15:01] intelligence. You know, on the one hand,

[15:03] it can solve gold medal problems in IMO,

[15:07] which is super hard, but on the other

[15:08] hand, as we've all seen, it can still

[15:10] make basic elementary maths errors if

[15:13] you pose the question in a certain way,

[15:16] right? So, or elementary reasoning

[15:18] errors. So, there's just something to me

[15:19] about the almost an introspection about

[15:22] its own thought process that I feel like

[15:24] there's there's something maybe missing

[15:26] there. Agents are really big. Some would

[15:28] say they're hyped. I personally think

[15:29] they're just getting started. It's

[15:31] [laughter] totally insane. What does

[15:33] DeepMind's internal research tell you

[15:34] about where agent capabilities actually

[15:36] are right now versus, you know, sort of

[15:38] the hype out there? I think we are I

[15:40] agree with you. I think we're just at

[15:42] the beginning. You have to have an

[15:43] active system that can actively solve

[15:46] problems for you to get to AGI. That was

[15:48] always clear to us. So, agents are that

[15:51] path and I think we're just getting

[15:52] going. I think all of us are getting

[15:54] used to how do we best work and you're

[15:56] leading the way in a lot of this in your

[15:58] own personal experiments. I'm sure many

[15:59] of you are doing that. I think how do

[16:01] you incorporate it into your

[16:03] workflow in a way that isn't just sort

[16:06] of a nice to have, but actually starting

[16:09] to do fundamental things. My I think My

[16:10] impression is at the moment we're all

[16:11] experimenting we're experimenting a lot

[16:13] of things, but we're only in the maybe

[16:15] the last couple of months starting to

[16:16] find the really valuable places. And the

[16:19] technology probably only getting good

[16:21] enough for that to be the case, right?

[16:23] Where that it's not a kind of toy nice

[16:25] demonstration, but actually really

[16:27] adding value to your to your to your

[16:29] time and efficiency.

[16:31] I had often wondered I see a lot of

[16:33] people working on

[16:34] like setting off, you know, dozens of

[16:37] agents for like 40 hours, but I'm not

[16:39] sure I've seen the output that yet of

[16:42] that quite justify that level of input

[16:45] going in, but I think it will come. So,

[16:47] I still think we're in the

[16:48] experimentation phase. We haven't seen a

[16:50] triple-A game that tops the App Store

[16:53] charts that was sort of vibe coded yet,

[16:56] right? I've seen and I've programmed and

[16:58] I'm sure many we've all done little nice

[16:59] demonstrations and it's like amazing. I

[17:01] can do a prototype of theme park in half

[17:04] an hour now which took me six months

[17:06] back when I was 17. It's kind of

[17:08] mind-blowing and I and I wish I I got

[17:10] this feeling if I spent the whole summer

[17:12] working on it, you could make something

[17:14] really incredible, but it still needs

[17:16] craft and, you know, human sort of soul

[17:19] into it and taste. I think that's that's

[17:21] something that can that's you have to

[17:23] make sure you still bring that to to

[17:26] whatever it is you're building. And I

[17:27] think it still shows like it's not quite

[17:29] there yet because why haven't we seen

[17:32] a kid making a hit game that's that

[17:34] sells 10 million copies, right? That

[17:36] should be possible given the effort

[17:38] that's gone in. So something's still

[17:40] somehow missing. Maybe it's to do with

[17:42] the process, or maybe it's to do with

[17:43] the tools. I'm not quite sure. You will

[17:45] probably know better than me cuz I'm

[17:46] sure you're all experimenting on that.

[17:48] But I haven't seen the result yet which

[17:50] I would expect once this is really

[17:53] delivering that full value. Which I

[17:55] think will come in the next 6 to 12

[17:57] months. Some of it is like how much of

[17:58] it will be autonomous versus I mean, I

[18:00] don't think we'd see autonomous first.

[18:02] We would actually probably see people in

[18:04] this room operating at 1000X, and then

[18:07] That's what you should see first, and

[18:09] then many of you, you know, they'll be

[18:11] like games companies or, you know, other

[18:14] types of companies that have built some

[18:16] kind of best-selling app, best-selling

[18:18] game using these tools. That's what you

[18:21] should see first, and then more of that

[18:23] will get automated. I mean, some of it

[18:25] is like there's a human in there, and

[18:27] then the human doesn't want to say that

[18:29] the the the agents did it yet. I think

[18:32] part of it might be though that um this

[18:35] if we want to discuss like creativity,

[18:37] what I often say about that is like if

[18:39] we look at the things we've done like

[18:41] AlphaGo. So obviously very famously

[18:44] you'll all know about the move 37 in

[18:45] game two, and for me I was waiting for a

[18:47] moment like that to start the science

[18:49] projects like AlphaFold. We started

[18:51] AlphaFold like the day we got back from

[18:53] Seoul, which is 10 years ago now. I'm

[18:55] going to Korea after this to celebrate

[18:58] the 10-year anniversary of AlphaGo. But

[19:01] it's not enough to come up with move 37.

[19:03] Like that's pretty cool, very useful,

[19:06] um but can it invent go?

[19:08] That's what I want a system that can

[19:10] invent go if you give it a high-level

[19:12] description, you know, like a game you

[19:14] can learn the rules of in 5 minutes, but

[19:17] it takes a many lifetimes to master.

[19:19] It's beautiful aesthetically,

[19:22] um but you can play it in a few hours in

[19:24] an afternoon. So, you know, maybe you

[19:26] could imagine that would be the

[19:27] high-level description I would give and

[19:29] then I'd want the the return the thing I

[19:31] get back is go.

[19:33] Right? And um clearly today's systems, I

[19:36] think can't do that. So, the question is

[19:39] why? Um and I think there's something

[19:41] still missing there. Well, someone in

[19:43] this room might might make it.

[19:44] >> Then the answer would be there's nothing

[19:45] missing. It just was the way we were

[19:47] using the systems. And that might

[19:49] actually be the answer. It might be that

[19:51] today's systems are capable of that with

[19:53] a brilliant enough creative person using

[19:56] it and providing that impetus that the

[19:59] soul of the project and being able to

[20:01] probably being

[20:03] au fait enough with the tools to like

[20:06] almost be at one with the tools. I could

[20:07] imagine that would be happening if you

[20:09] experimented with the tools all day and

[20:11] all night like probably many of you are

[20:12] doing that and you combine that with

[20:14] proper deep creativity.

[20:17] Um something, you know, more incredible

[20:18] could be done. Switching gears to open

[20:20] source, I mean or open open and open

[20:22] weights. I mean, the recent release of

[20:24] Gemma, you're making highly capable open

[20:28] and accessible ones that can actually

[20:29] run locally. What do you think that

[20:31] means for you will AI be something that

[20:35] is in the hands of the users instead of

[20:36] primarily in the cloud? And does that

[20:39] change who gets to, you know, build with

[20:41] these models? We're huge proponents of

[20:44] in general of open source and open

[20:46] science. And you mentioned AlphaFold at

[20:48] the beginning, you know, we put that all

[20:49] out there for free. And all of our

[20:51] science work even still today we publish

[20:54] in, you know, the big journals. We

[20:56] wanted to create uh world-leading models

[20:58] for their their sizes. Right? And so,

[21:00] that's what we hopefully we've done with

[21:02] Gemma. And we're, you know, very

[21:03] committed to that path. And hopefully

[21:05] you will experiment and build and and

[21:07] enjoy and using Gemini. I think it's

[21:08] been like 40 million downloads now and

[21:11] uh it's just in you know 2 and 1/2

[21:13] weeks. So we're really excited about

[21:14] that. And I also think it's important

[21:16] for there to be Western stacks on open

[21:19] source. You know, obviously a lot of the

[21:20] Chinese models are excellent and and

[21:23] they're currently well well leading in

[21:24] open source and we think Gemini is very

[21:26] competitive for its sizes

[21:28] uh in in all those respects. And for us,

[21:31] I mean there is a question of resources,

[21:33] talent, and compute. Like nobody has

[21:35] enough spare compute to just make two,

[21:38] you know, uh frontier models at maximum

[21:41] size, right? With different attributes.

[21:43] So that's pretty difficult. But also for

[21:45] what for now what we've we've decided is

[21:47] that our edge models, the things we want

[21:49] to use for Android and glasses and

[21:52] robotics, um it's best that they're open

[21:54] models because they're vulnerable anyway

[21:57] on the once you put them out on the

[21:58] surfaces. So they might as well be

[22:00] actually fully open, right? So we've

[22:03] sort of made a decision to kind of unify

[22:05] that

[22:06] uh at the at the kind of we call it nano

[22:09] size level. So that actually works for

[22:11] us uh strategically as well. Um and you

[22:15] know, we hope as many people as possible

[22:16] build on it. And of course, we'll be

[22:18] building on that, too. Earlier uh before

[22:20] we came on, I got to show you a demo of

[22:22] uh my version of Samantha from Her,

[22:24] which is Yes. uh harrowing for me to try

[22:26] to demo something to you. Yeah, very

[22:28] good. Um and it worked, which is

[22:30] amazing. Gemini was built multimodal and

[22:32] I spent a lot of time with a bunch of

[22:34] the models and I mean the depth of the

[22:36] context and the tool use with speech

[22:40] directly to model, like there's nothing

[22:42] like bar none, like the best one

[22:44] actually.

[22:44] >> Yeah. Yeah, I think I think that's the

[22:46] sort of still a slightly

[22:47] underappreciated aspect of of of the

[22:49] Gemini series is we we started it being

[22:52] multimodal from the start. That made it

[22:54] a little bit more difficult actually to

[22:56] begin with cuz then just focusing on

[22:57] text, for example. But I we believe

[23:00] we're going to gain from that in the

[23:02] long run. And I think we're seeing that

[23:03] now for

[23:05] things like world model building, so

[23:08] stuff like Genie that we build on top of

[23:10] Gemini. I think it's going to be really

[23:12] important for things like robotics. So

[23:14] this is why Gemini robotics which many

[23:15] of you probably played around with, I

[23:17] think it's going to be built on

[23:18] multimodal foundation models, the

[23:20] robotics models. And we think we have a

[23:23] sort of competitive advantage with with

[23:25] Gemini being so strong at multimodal.

[23:27] We're using it increasingly in things

[23:29] like Waymo. Um but also if you imagine

[23:32] devices and assistants that digital

[23:35] assistants that come with you into the

[23:36] real world, you know, maybe on your

[23:38] phone or glasses or some other device,

[23:41] it needs to understand the physical

[23:43] world around you and intuitive physics

[23:46] and and the and the physical context

[23:48] you're in. And that's what our systems

[23:50] are extremely good at and I think you

[23:52] found that's why you've enjoyed using it

[23:53] in your setup. We're planning to

[23:55] continue on that and I think we're

[23:57] far and away the strongest models on on

[23:59] those types of

[24:01] problems. So the cost of inference is

[24:03] dropping fast. What becomes possible

[24:05] when inference is essentially free and

[24:07] how does that change what your team is

[24:09] actually optimizing for? Yeah, I'm not

[24:11] sure inference

[24:13] will ever be essentially free. I mean

[24:15] there's sort of Jevons' paradox and

[24:17] other things about like I think we'll

[24:18] just end up using all of us will end up

[24:21] using whatever we can get our hands on

[24:24] and you could imagine

[24:26] millions of agents, swarms of agents

[24:28] working together on things. So that's

[24:30] one way to use the inference or you

[24:32] could imagine

[24:33] single agents or smaller groups of

[24:35] agents thinking for in multiple

[24:38] directions and then ensembling that. So

[24:40] we're experimenting with all these

[24:41] things, probably many of you are. All of

[24:43] that will use up any inference I think

[24:46] that's available. I mean one day maybe

[24:48] it can be almost cost zero, certainly

[24:51] the energy if we solve fusion or you

[24:53] know, superconductors or you know,

[24:54] optimal batteries or some set of those

[24:56] things which I think we will do with

[24:58] material science, And energy costs will

[25:00] be essentially zero, but there'll still

[25:02] be the physical creation of the chips

[25:04] and other things. There'll There'll be

[25:06] some bottleneck um at least for the next

[25:09] few decades, I think. And so if that's

[25:11] the case, there'll still be rationing on

[25:14] the inference side. You still have to

[25:16] use it, I think, efficiently. Yeah.

[25:18] Well, luckily the smaller models are

[25:19] getting smarter and smarter, which is

[25:20] fantastic. Uh we got a lot of bio and

[25:23] biotech founders in the audience. I can

[25:26] see a few. AlphaFold 3 took us beyond

[25:28] proteins to a broad spectrum of

[25:29] biomolecules. Uh how close are we to

[25:32] modeling full cellular systems, or is

[25:34] that still a fundamentally harder

[25:36] problem in a class of its own? Well, I

[25:39] Isomorphic Labs, which we spun out from

[25:41] from from from DeepMind after we did

[25:44] AlphaFold 2,

[25:45] um it's it's which is going amazingly

[25:47] well. It's it's it's trying to build out

[25:49] uh not just AlphaFold. It's just one

[25:51] piece of the drug discovery process, uh

[25:54] as many you know, but we're trying to do

[25:56] the the adjacent biochemistry and

[25:58] chemistry to design the right compounds

[26:00] with the right properties and so on.

[26:02] We'll have some big announcements for

[26:03] you know, very soon to talk about on the

[26:05] on that front. I think that's going

[26:07] really well. Eventually, you want a

[26:09] whole virtual cell. So I've talked about

[26:11] this in many of my science talks about a

[26:13] full working simulation of a cell that

[26:16] you can perturb, and then the you know,

[26:18] the the outputs of that would be close

[26:21] enough to experimental that it's useful,

[26:23] right? You could skip out a lot of the

[26:25] the search steps and generate lots of

[26:27] synthetic data to train other models

[26:30] that then would predict things about,

[26:31] you know, real cells. And um

[26:34] I think we're about 10 years away

[26:36] probably from something like a virtual

[26:37] cell, like a full virtual cell. You

[26:39] know, we're starting out This is we're

[26:41] working on the DeepMind side, science

[26:42] side, on a you know, virtual nucleus,

[26:45] cell nucleus first cuz relatively

[26:47] self-contained. The trick with all of

[26:49] these things is can you pick uh a slice

[26:51] of the complexity, you know, eventually

[26:53] you want to want to model a human body,

[26:55] but can you model it down to the right

[26:57] level of detail and what slice can you

[27:01] take out of it that will be

[27:02] self-contained enough? You can kind of

[27:04] model and approximate the inputs and

[27:07] outputs into that self-contained system

[27:09] and then just focus on the

[27:10] self-contained system. So, a nucleus is

[27:12] quite interesting from that perspective.

[27:15] Um, then the other issue is just there's

[27:16] not enough data yet. So, you need data

[27:20] and I talked to various, you know, top

[27:22] scientists about who work on electron

[27:24] microscopes and other imaging things. If

[27:27] we could image a live cell without

[27:29] killing the cell, that would be

[27:32] game-changing obviously cuz then you

[27:33] could convert it into a vision problem

[27:35] which we would know how to solve. Right?

[27:37] And but at the moment, there are at

[27:39] least I'm not aware of any techniques

[27:41] that can give you a kind of, you know,

[27:42] nanometer resolution

[27:45] but without destroying but in, you know,

[27:48] in a live dynamic cell. So, you can see

[27:50] all the interactions. Right? You can

[27:51] take static images at that resolution

[27:53] obviously.

[27:55] Really detailed now and that's quite

[27:56] exciting but it's not enough to turn it

[27:59] just into just into a complex vision

[28:02] problem.

[28:03] So, that's one way it could be solved.

[28:05] So, it could be a hardware driven data

[28:06] driven solution or it could be that we

[28:09] build better

[28:11] learn simulators of these dynamical

[28:14] systems. So, that's that's the more

[28:15] modeling way of solving it. You've been

[28:18] looking at all kinds of science and not

[28:19] just bio. There's material science, drug

[28:22] discovery, climate modeling,

[28:23] mathematics. If you had a rank which

[28:26] scientific domain will transform the

[28:27] most dramatically the next 5 years,

[28:29] what's in your list?

[28:30] >> all sounds exciting and that's why, I

[28:32] mean, that that for me has been my main

[28:34] passion and always the reason why I've

[28:36] worked on AI for my whole career for 30

[28:38] plus years now is to use AI as the

[28:41] ultimate tool. I always thought AI would

[28:43] be the ultimate tool for science and to

[28:45] invite such advanced scientific

[28:48] scientific discovery, and things like

[28:49] medicine, and just our understanding of

[28:51] the universe around us. So, actually,

[28:53] when you mentioned our original way we

[28:55] used to articulate our mission

[28:56] statement, which is still the way we

[28:58] think about it, is there was two steps

[28:59] to it. One was Step one was solve

[29:01] intelligence, i.e. build AGI, and then

[29:03] step two was use it to solve everything

[29:05] else. We had to change that a bit over

[29:07] time cuz people were like, "Do you

[29:08] really mean solve everything else?" And

[29:10] we did mean that, and I think people are

[29:12] sort of understanding what that means

[29:14] today. But, specifically, I was meaning

[29:16] solve other what I call root node

[29:18] problems in science. So, areas of

[29:20] science that would unlock whole new

[29:22] branches or avenues of discovery. And

[29:24] AlphaFold is the prototypical example of

[29:26] what we want to do. So, over 3 million

[29:28] researchers around the world, pretty

[29:30] much every biology researcher in the

[29:31] world uses AlphaFold now. And I was told

[29:34] by some of my, you know, former

[29:36] executive friends that, you know, almost

[29:39] every drug discovered from now on will

[29:42] have used AlphaFold at some point in its

[29:44] in the drug discovery process. So,

[29:46] that's something we're very proud of,

[29:48] and it's the sort of impact that we hope

[29:50] to have with with AI. But, I do think

[29:52] it's just the beginning. I I don't

[29:54] really see any area of science or

[29:56] engineering that this won't be able to

[29:57] help be helpful with. And the ones you

[29:59] mentioned, I think we're almost like an

[30:02] AlphaFold one moment. So, it's we've got

[30:04] very promising results, but it's not

[30:05] quite solved the the grand challenge yet

[30:08] in that domain. But, I think we're going

[30:10] to have a lot to talk about in the next

[30:12] couple of years on all those areas you

[30:13] mentioned, materials, which I I think is

[30:15] very exciting, all the way to

[30:17] mathematics. In in science, I mean, it

[30:19] feels Promethean. It's like, here is

[30:21] this capability, and you I think so. I

[30:24] mean, of course, along with that,

[30:25] including including what the the the

[30:27] parable of Prometheus, we have to also

[30:29] be careful with how we use that and what

[30:32] we use it for, and also the misuse that

[30:35] can happen with those same tools. A lot

[30:37] of people in this room are trying to

[30:37] build companies applying AI to science.

[30:40] For them, what's the difference between

[30:41] a startup that actually advances the

[30:43] frontier in your view versus one that's

[30:45] just wrapping an API around a foundation

[30:47] model and calling it AI for science?

[30:49] Well, look, I think there's one of the

[30:51] things I would recommend. I'm trying to

[30:52] think about and I think you mentioned

[30:53] this to me before. What would I do today

[30:55] myself if I was sitting in your place in

[30:58] Y Combinator, you know, looking at

[30:59] things. One thing you have to do is

[31:01] obviously intercept where the AI tech is

[31:04] going. So, that's one hard part of it.

[31:06] But, I do think there's huge scope for

[31:09] combining where AI is going with some

[31:12] other deep technology area. I just think

[31:14] that that sweet spot is is whether it's

[31:16] materials or medicine or other really

[31:18] hard areas of science. I think that

[31:21] those kinds of interdisciplinary teams,

[31:24] especially if it involves the world of

[31:25] atoms as well,

[31:27] there's not going to be a shortcut to

[31:28] that, at least in the foreseeable

[31:30] future. Those areas that are pretty safe

[31:33] from just getting swamped by whatever

[31:35] the next update is to the foundation

[31:37] models. So, I think if you're looking

[31:39] for things like that, that's one of the

[31:41] more defensible areas I would say. And

[31:43] I've always loved deep tech, so I'm kind

[31:45] of biased towards deep tech things. I

[31:48] think nothing

[31:49] that's really long-lasting and

[31:51] worthwhile is easy. And so, I'm always

[31:54] been drawn to to deep technologies.

[31:57] Obviously, AI was like that back in 2010

[31:59] when we started out, right? It was It

[32:01] was thought to just we know we know it

[32:03] doesn't work kind of thing is what I was

[32:05] told by investors and even in academia

[32:07] it was considered to be a very niche

[32:10] subject that we sort of tried in the

[32:12] '90s and we know doesn't work. But, if

[32:14] you, you know, if you have belief and

[32:16] conviction in your idea why it's

[32:18] different this time or what special

[32:19] combination from your background that

[32:22] you had, ideally you're expert in both

[32:24] those areas, both the machine learning

[32:26] and the other area you're applying it to

[32:27] or you can create a founding team with

[32:29] that expertise, I think there's huge

[32:31] impact to be made there and huge value

[32:33] to be built there. That's of important

[32:35] message. I mean, even I mean, it's hard

[32:38] it's easy to forget. Like, basically,

[32:39] once you've done it, you've done it.

[32:40] But, before you've done it, people are

[32:42] arrayed against you. Oh, sure. I mean,

[32:44] no one believes in it, which is why I

[32:45] think you got to you've also got to work

[32:47] in things that you're genuinely

[32:49] passionate about. Like, for me, I would

[32:52] have worked on AI no matter what

[32:54] happened. I just decided from a very

[32:56] young age it was the thing that um could

[32:59] be the most consequential thing I could

[33:01] think of. It's turned out that way, but

[33:02] it might not have. Maybe we would have

[33:03] been 50 years too early. And it was also

[33:06] the most interesting thing I could think

[33:08] of working on. And so, I would have

[33:10] still be working on AI today even if we

[33:13] were still, you know, in a little garage

[33:15] somewhere and it still wasn't quite

[33:16] working. I would have still been trying

[33:18] to find Maybe I'd have been back in

[33:19] academia or something, but I would have

[33:20] found some way of of continuing to work

[33:23] on it. So, I mean, AlphaFold was like an

[33:24] example of a spike that you pursued and

[33:27] it worked. You know, what makes a

[33:29] scientific domain ripe for an AlphaFold

[33:31] style breakthrough? And is there a

[33:32] pattern, a certain objective function?

[33:35] >> The way I I I I should write this up at

[33:37] some point when I have 5 minutes spare,

[33:39] but the lesson I've learned from all the

[33:42] Alpha projects we've done, specifically

[33:44] AlphaGo and AlphaFold, is um I think the

[33:48] techniques we have and the problems I

[33:49] look like to look for are great in if

[33:52] this if the situation can be described

[33:54] as massive combinatorial search space.

[33:56] The more massive the better in some

[33:58] ways. So, no brute force or special case

[34:00] algorithm will will solve it. And that's

[34:02] true of Go moves and of, you know,

[34:04] different configurations of proteins,

[34:07] far more than the atoms in the universe,

[34:08] both of those. And then, um you have a

[34:11] clear objective function. So, you know,

[34:13] you can think of it as minimizing the

[34:14] free energy in the proteins or, you

[34:16] know, the winning the game of Go. So,

[34:18] you need to be able to you need to

[34:19] specify your objective function clearly

[34:21] so you can hill climb. And then, um

[34:24] enough data and or simulator that can

[34:27] generate you uh lots of uh in

[34:29] distribution

[34:31] uh uh synthetic data. If those things

[34:33] are true, then I think um with today's

[34:36] methods, you can go a long way into

[34:38] tackling and finding the kind of needle

[34:40] in the haystack that you need uh to for

[34:42] the solution that you're trying to look

[34:44] for. And I think of just drug discovery,

[34:45] by the way, in the same way, right?

[34:47] There is a compound out there that would

[34:49] solve this disease if one could find it,

[34:52] if one could only find it, right? And

[34:54] that wouldn't have any side effects and

[34:55] so on. And as long as the laws of

[34:57] physics allows it, then the only

[35:00] question is how do you find it in an

[35:01] efficient way, in a tractable way? I

[35:04] think we showed for the first time,

[35:05] actually, with AlphaGo, that these

[35:07] systems could uh find those kinds of

[35:10] needles in a haystack, in that case, you

[35:12] know, the perfect Go move. I guess uh to

[35:14] get a little meta, I mean, we're we're

[35:15] talking about humans using these methods

[35:18] to create AlphaFold, but then there's a

[35:20] meta level, which is humans using AI to

[35:23] explore the space of possible

[35:25] hypotheses. How close are we to AI

[35:27] systems that can do genuine scientific

[35:29] reasoning, not just pattern matching on

[35:32] data?

[35:32] >> we're close. Um

[35:35] we're working on these general systems

[35:36] like that like I think we we have this

[35:38] system called co-scientist, and we have

[35:41] other algorithms like AlphaFold that can

[35:43] go a little bit beyond what the basic

[35:45] Gemini will do. And obviously, all the

[35:47] frontier labs are experimenting in this

[35:49] way. I've yet to seen anything so far,

[35:52] and we we all tinker with the same

[35:53] things, you know, some math problems

[35:54] that are a little bit harder than IMO

[35:56] and so on. I haven't seen anything yet

[35:59] um that is a true genuine, you know,

[36:02] massive discovery. That's my personal

[36:04] opinion. I think it's coming. I think it

[36:07] may be related to uh this earlier

[36:10] this thing we discussed about

[36:11] creativity, and and actually going on

[36:14] beyond the bounds of what's known. So,

[36:16] clearly, that's just not pattern

[36:17] matching at that point, cuz there is no

[36:19] pattern to match to, and it's a bit more

[36:21] than extrapolation. It's some kind of

[36:23] analogical reasoning, and I don't think

[36:25] these systems have that, or at least

[36:27] we're not using them in the in the right

[36:29] way to do that. So, the way I often say

[36:31] that in science is can it come up with a

[36:33] hypothesis that's really interesting,

[36:36] not just solve one. When I say just,

[36:38] we're not talking about just like

[36:39] solving the Riemann hypothesis or

[36:41] something. This would be obviously

[36:42] amazing, or one of the Millennium Prize

[36:44] problems, and maybe we're a couple of

[36:46] years out from doing that. Um but, I'd

[36:48] like to solve P equals NP. That's That's

[36:50] my favorite one. But, can you But, even

[36:52] harder than that would be to come up

[36:54] with a new set of of Millennium Prize

[36:56] problems that were regarded by top

[36:59] mathematicians to be as, you know, deep

[37:01] and meaningful and worthy of lifetime of

[37:04] study and effort to solve. Right? I

[37:07] think that's another level harder. And

[37:10] uh we don't have um you know, I still

[37:12] don't think we know how to do that. I

[37:14] don't think it's it's magical, though. I

[37:16] do think these systems will be

[37:18] eventually be able to do that. Maybe

[37:20] we're missing one or two things. And

[37:21] then, the way we would test that is, you

[37:23] know, I sometimes call it my Einstein

[37:25] test, which is, you know, can you train

[37:27] a system with the knowledge of cutoff of

[37:29] 1901, and then will it come up with you

[37:33] know, what Einstein did in 1905,

[37:34] including special relativity, you know,

[37:36] his annus mirabilis. Can Can it do that,

[37:39] right? Uh and then, I think we could run

[37:42] that test. May- Maybe we should just run

[37:44] that test and keep seeing if that's

[37:46] possible. And once that is, then I think

[37:48] we're on the verge of these systems

[37:49] being able to invent something new,

[37:51] truly novel. So, last last question. For

[37:54] the people who are deeply technical in

[37:55] this room who want to work on something,

[37:59] you know, even close to the scale that

[38:01] what you have created with you know,

[38:02] it's one of the largest AI efforts in

[38:04] the world, and you've been a pioneer for

[38:06] all these years. So, for that, I think

[38:08] everyone in this room thanks you and the

[38:10] folks at DeepMind very, very deeply from

[38:12] the bottom of our hearts. Thank you.

[38:14] What's the thing that you know now about

[38:16] building at the frontier that you wish

[38:17] you'd known at 25?

[38:20] I think we covered some of it in terms

[38:22] of actually you you work out that going

[38:24] after hard problems and deep problems um

[38:28] it's no more difficult in some ways than

[38:29] than going after a shallower, simpler,

[38:32] more superficial problem. They're

[38:33] they're they're just differently

[38:34] difficult. There's different things that

[38:36] are hard about each of those things, but

[38:38] I think given life's very short and you

[38:41] you know, you only have so much time and

[38:43] energy, you might as well put your life

[38:45] force into something that will really

[38:47] make a

[38:48] difference if you hadn't done it, if you

[38:50] hadn't been there to push it. So, I

[38:52] would just think of it through that

[38:54] lens. And then the other thing is if

[38:56] you're if you are and then we talked

[38:57] about deep tech and I love

[38:59] interdisciplinary

[39:00] uh work and I think that's going to be

[39:02] even more prevalent in the next few

[39:04] years in combinations of fields and uh

[39:07] uh finding the the the the connections

[39:09] between those fields. And it's going to

[39:11] be even easier to do that with AI. And

[39:13] then the only other thing I would say is

[39:15] if you know, if you have your depending

[39:17] on what your AGI timeline is, you know,

[39:19] mine's like 20 30 or something like

[39:20] this, then if you start off on a deep

[39:23] tech journey today

[39:26] usually that you're talking about a

[39:27] 10-year journey for for true deep tech

[39:30] in my opinion. So, then now you have to

[39:32] just consider AGI appearing in the

[39:35] middle of that journey. So, what does

[39:37] that mean? It doesn't it's not bad

[39:38] necessarily, but you have to take that

[39:40] into account, right? To will it be able

[39:43] to leverage it? What will the AGI system

[39:45] do with it? And it goes a little bit

[39:47] back to what you said earlier about

[39:48] AlphaFold and general AI systems. So,

[39:51] one thing I can think see happening is

[39:53] Gemini, Claude, or one of these general

[39:55] systems making use of AlphaFold like

[39:58] specialized systems as tools. I don't

[40:00] think we're going to have it just in one

[40:02] giant brain cuz it will have too much

[40:04] regression in if I put all the proteins

[40:07] into you know,

[40:08] Gemini, that wouldn't make sense. We

[40:10] don't need Gemini to do protein folding.

[40:12] Going back to your information

[40:13] efficiency, it will definitely affect

[40:15] its language skills or something like

[40:16] that, right? In a bad way. So, much

[40:19] better I think is to have really good

[40:21] general purpose tool usage models that

[40:23] will then

[40:24] maybe they could even train those

[40:26] specific tools, but they would be in a

[40:27] separate

[40:29] system. So, I think that's kind of

[40:31] interesting to think through the

[40:32] implications of that and then what you

[40:33] might build today. Also, physical things

[40:36] too like what kinds of factories would

[40:37] you build, what sorts of

[40:40] you know, finance systems and so on. So,

[40:42] I just think you need to really take

[40:44] that seriously and and and on the one

[40:46] hand is like an imagine what that world

[40:47] would look like and then build something

[40:49] that would be useful if that comes in

[40:51] halfway through.

[40:53] Demis Hassabis everyone.

[40:54] >> [applause]