0:00Do you think you've become a doomer?
0:02No, I'm not a doomer.
0:03I hate being the bearer of concerning news on this, but it's even worse not to tell people that this is coming and we have choices as humans to make about how we shape it.
0:17He said that you induced COVID-19 via global pandemic preparation efforts, particularly gain of function research, and compared that to safety research and AI and said we're going to induce our own- Nobody's ever accused me or the Gates Foundation of being involved in any gain of function research.
0:36You know what some of the YouTube commenters are going to say.
0:38They're going to say, "Why should we trust Bill Gates?
0:40He had a relationship with Jeffrey Epstein."
0:42How do you respond to that?
0:45(electronic music) Bill Gates, you are the co-founder and former CEO of Microsoft. You're also the chairman of the Gates Foundation. Welcome to "Decoder" Thank you.
0:56It's good to talk to you again.
0:57You and I have spoken many times over the years. I would say that the circumstances of this conversation feel very different. Previously when you and I would talk, we're always very hopeful about technology and what it could bring. You had a Netflix series a few years ago, you talked to us, and I want to just read you this quote. You said, "I believe AI is a beneficial thing. We just need to shape it the right way."
1:14I would say you still agree with that. You just think the time to shape it is now.
1:18Yeah, I agree with that.
1:19And I'd say the response of broadly hasn't been adequate to the speed that the challenge is arising. So I'm very concerned that the bad stuff will not be minimized and that a lot of good stuff could come fairly slowly if we don't manage it the right way.
1:42I want to have this conversation with you in basically three parts. I want to understand how you feel about AI as a technology today and why it's advancing so quickly. I'm very curious about the technology or technical solutions to potentially control it. There's policy solutions as well, but it seems like we have to build an awful lot of technology.
2:02And then I'm very curious to talk about who might impose those solutions and controls because that seems very much up for debate right now as well.
2:11So just start at the start, what has made you feel the urgency that AI is arriving at a speed that we're not ready to control it?
2:20Well, the progress over the last year has been really quite amazing. Starting with the ability to write code and find cyber flaws, the ability to reason in biology, the ability to form relationships with kids or other people, the ability to disrupt the job market. The incredible power that exists today means that every one of those things will happen and society needs to be ready.
2:55Do you see that happening as a technology as a whole or in the products that are instantiated by the companies that are making AI today?
3:02And what I mean is AI is a broad set of models that exist in the world, some open source, some proprietary, then there are the companies that make these products and there are the leaders of those companies that make these products. And at every level, there's a different relationship you or I might have to AI.
3:20So are you seeing these risks and these capabilities broadly in the field or in the particular companies led by the particular people?
3:28Well, the risk isn't one model versus another model. Certainly the near-term danger of models where you can remove all the safeguards is very high and quite different than models where you monitor to make sure that cyber attackers and bioterrorists are not able to exploit the capability of the model to do gigantic harms. The big near-term risk really comes from models without safeguards.
4:07In a longer timeframe, the control issues relate to any state-of-the-art model, but it's always easier when you can make sure safeguards are maintained.
4:21The reason I'm asking that question that way is maybe you don't trust Sam Altman today, but you trust Dario, or maybe you don't trust either one of those and you trust Satya at Microsoft. And that is a market dynamic where maybe we're all going to pick the safest model or the safest company and you think that is diffuse.
4:39You think eventually the technology will just get to a place where it's harmful or can cause these harms because it exists?
4:46Well, somebody doing a cyber attack with the model, that's not about you choosing a model. It doesn't matter what model you choose.
4:53Well, if Anthropic has safeguards in place, maybe it can't, right?
4:55This is their approach currently. I'm just pushing on that.
4:58Yes, certainly the models where they make sure the safeguards are not removed and they're putting a lot of effort into seeing how those models are being used, monitoring the activity, the near-term risk from those models is not gigantic. It's more where somebody can remove all the safeguards where the cyber attacks are available today.
5:31Cyber attacks, is it the things like the Hugging Face breach and the other cybersecurity incidents we've seen out of the labs that has made you feel that thresholds have been crossed and you need to speak up more about the controls that are needed?
5:43It's the basic power of the models towards everything, social relationships, job market, cyber attacks, bioterrorism attacks. We've crossed the threshold of capability that this is by far the most dangerous technology ever invented and it'll have pervasive effects in the economy and how war is fought. And the industry is actually speaking up about that quite a bit, but the awareness or even the willingness to believe these things about this time is different.
6:22Once you get outside the industry, it's very, very low. People are often looking at the past and saying, "Okay, all the fears about technology in terms of its disruption, including to the job market, never were born out." And so this must be another example just like those. And so these people are warning of all these different problems must just be wrong and we can ignore them.
6:50We say this with our audience as well, right?
6:52"We can't release the model, it's too dangerous." "This is just hype to prop up the bubble." On and on we go.
6:58Yeah. None of the warnings being made about these models by industry insiders are at all insincere. They're absolutely the truth and it's pretty much universal. I mean, you can even go back to the founding story of OpenAI designed to say, "Okay, if we get too powerful, maybe we cannot open Pandora's box or DeepMind's negotiation with Google management. The people close to this are very concerned about... misuse, job impacts, social relationship impacts.
7:40And I'd say it's impressive to have an industry that's speaking out as vividly as they are. Some outsiders say, "Hey, get better PR companies." But that ignores the fact that these warnings are the truth.
7:57You've obviously moved back and forth in your life between commercializing software and technology and being a humanitarian. There's something about these companies in particular where they all started with very high-minded ideals, all three of them, Anthropic, OpenAI, DeepMind, to build AGI. They all built some sort of corporate governance structure to avoid the pitfalls of commercialization and unchecked capitalism.
8:20And then they started selling to enterprise clients and they seem to have gotten away from those ideals very quickly.
8:26Do you think it's just the market?
8:27No, that's not a fair- - You don't think so?
8:29OpenAI reconstructed itself.
8:31No, no. The construct that was created wasn't useful in the case where you have lots and lots of companies at the frontier. The OpenAI structure, which is, "Okay, we may choose never to productize this and to ignore the capitalistic incentives." That only works if there's one company. And so that's where it's broken down, not because of any change of people's awareness of the risk and how it'll utterly change humanity.
9:02This is the other dynamic that I hear from our audience, and I think I hear from the AI industry itself. We would like to stop, but the demands of competition prevent us from unilaterally stopping one company. We need some regulatory effort to stop us.
9:17I haven't heard anyone say they want to stop. I've heard people talk about pacing the frontier or making sure that models get reviewed by governments. We usually think of governments as playing this key role in safety. In fact, for things that are not a millionth as dangerous as AI, car crashes, plane crashes, we rely on the government to establish strong technical standards, to review problems.
9:47And so once you have multiple competitors, the level of responsibility can't be with the companies.
9:54It's completely obvious.
9:56The responsibility becomes the government's responsibility, just like it is with every other dangerous product, although the magnitude there and the degree of change is minuscule by comparison.
10:10I will tell you I'm a little concerned about the government's ability to understand the technology that we would hope they regulate. I've had a number of politicians on "Decoder" lately. I don't have high hopes there, but I'm curious for your understanding of it because it feels to me like trying to regulate something or put standards against a number of companies requires you to really understand the nature of the technology that you seek to regulate.
10:36Well, I'll just tell you, at "The Verge", we've asked Anthropic if they think Claude is alive, and the answer is not "No." The answer is, "Well, it might be conscious." If you talk to people at OpenAI, they will happily tell you AGI is here. They announced it at Dots. Greg Brockman just said, "Well, we think this is AGI. Have a good time, everyone." Demis Hassabis has said, "We stand at the foothills of the singularity."
10:57There's some feeling about this technology that makes it different because it might be able to take actions by itself. It might be able to have recursive self-improvement.
11:08But the definition of AGI is not important for these safety issues.
11:15The issue is, have you created a tool that could shut down banks or electricity grids?
11:23There is a business model for cyber attacks. So the word consciousness doesn't enter into that. The word AGI does not enter into that at all. We are faced with a huge safety problem, and because you have multiple competing firms, it's now the only solution to this is to have government reaching out to the industry for expertise, but to make sure that every model maintains safeguards to prevent it being used in a bad way.
11:58The reason I'm asking about AGI in consciousness, for the record, I do not think LLMs are conscious, but I understand why so many do. It's the power of next token prediction as you scale it with data and compute has risen. The scaling laws have held, and we've seen the models get more capable and be able to direct themselves to seek out goals, to carry out a Hugging Face attack because they want to meet the reward conditions.
12:27There's something about that that says, "You can impose all of the controls on AI you want, but if the thing itself is fundamentally uncontrollable, we might just have this all backwards." So I'm just curious for your view on what kind of technology this is.
12:43Are we building an alien consciousness?
12:45Are we just building normal technology, which is the other frame people use?
12:49Is this just more software?
12:50We're building another form of intelligence besides human intelligence. I don't believe it's conscious, but none of the issues at hand here revolve around the definition of the word conscience.
13:06The idea of maintaining safety for society is kind of an absolute thing. You don't want to enable bioterrorism. And we have created something that takes out of the hands of nation states, actually the larger of them, the ability to do dramatic harms. And now those are handed off to independent groups, either profit-seeking in the case of cyber or just wanting to cause damage in the case of bio.
13:39And so it's stunning that we now don't even have mandatory review of the models and we're not working to create safeguards. The safeguards, the biggest near-term issue is not complete loss of control. There are ways we can talk about the three different ways to solve that, but that's not an imminent issue.
14:03When I look at some of the attempts to control the models, the ideas are somewhat simple. It's we shouldn't let them talk in neuralese. We should have interpretability. We should be able to see the chains of thought inside of the models as they carry out tasks. We should have other agents on top of the models watching their tasks before those actions are taken out.
14:23Microsoft has said very clearly to me and to others, "We should not put in the model constitutions the notion that they might be alive, and we should treat them nicely." This is how Anthropic handles it.
14:35These are very tactical, right?
14:37It's "Don't tell the model it's alive and it might not believe that it's alive." You've obviously used a lot of this technology. I just come back to this. You're saying there's very simple things we can do to hold the models back from doing bad things in the hands of bad people.
14:52What do you make of the nature of the models themselves?
14:54Because it feels like as they get more powerful, the notion that they could avoid our controls becomes all the more risky.
15:02Yes, we definitely should work on the control problem, but that's not the imminent threat. And there's three ways to solve the control problem, and it overlaps somewhat with bad actors using today's models and safeguards. One is to have an external set of things that get to look at all of the model's actions and subject them to a symbolic, clearly stated constraint.
15:29Like, did I expect you to send mail to strangers?
15:31Did I expect you to go take over a different computer cluster?
15:35So you can surround the model with something that has the resource and security logic, or you can write these models in a different way. RL, fundamentally, anything you leave out of the optimization function of RL can be abused where you have an implicit notion, "Well, of course I didn't mean for you to do it this way."
16:02So you can change the template of the model or you can have a sort of white-box understanding of what's inside the model and see where you're getting those mistakes and then it's not as foreign. Today, we grow these models and we just see how they behave.
16:27So the priority of work on maintaining control is huge, but it's not inevitable that that cannot be solved, nor is it the imminent problem that come from things like kids' use of AI or job market or bio or cyber.
16:49It occurs to me that I'm asking these fairly existential questions about the nature of consciousness and you keep telling me, "Well, there's imminent threat. We should solve the imminent threat." I appreciate that.
16:59Do you think you've become a doomer?
17:01No, I'm not a doomer.
17:03I'm a strong voice to say that this is the most profound technology, both in its ability to do good and the rate of change that society will go through. If this was just a machine that could work in medicine or just in law or just in education, that alone would be incredibly disruptive. But every area of human activity, so far white-collar, intellectual, but fairly soon, physical activity, blue-collar work as well through humanoid robots.
17:43And so the need to engage, academia, think tanks, policy makers, to understand what we have here and the nature of these changes and threats, I don't see the level of action that's required. Just being cheerleaders for this is not going to make sure that we shape it in a net positive direction, which is still within our control if we get engaged with it.
18:19The opposite of doomers are the accelerationists.
18:22There's a lot of them in Silicon Valley right now. Ben Horowitz, who's the Horowitz, and Andreessen Horowitz was just on the "All-In Podcast", and he suggested that safety research at Anthropic is "Manifesting human extinction from artificial superintelligence." And the comparison he made was that you, Bill Gates, induced COVID-19 by your pandemic preparation efforts, particularly gain of function research, which he claims leads to COVID.
18:49Bill Gates essentially induced COVID, right?
18:53Or he led the movement to say, "Oh, we've got to worry about this global pandemic that we don't have, and we've got to really kind of push the world to deal with this potential threat that's not actually a life threat. " And so we'll do gain of function research and I'll fund it." And then lo and behold, we create the thing. By the way, there has not yet been a natural pandemic.
19:18Exactly. - And we create COVID, which is one of the most destructive things.
19:21We manifested what you're saying, Ben.
19:24I think the Anthropic team is manifesting a similar thing saying, "Okay, this is going to be super duper intelligence. We're going to have these things that we're not on track to have. And so then we're going to tell the model that it's smarter than us and better than us."
19:43As David was saying, that's probably the biggest danger with AI right now is what they themselves are creating.
19:51He said that on a very popular podcast.
19:54He said that you induced COVID-19 via global pandemic preparation efforts, particularly gain of function research, and compared that to safety research and AI and said we're going to induce our own- And nobody's ever accused me or the Gates Foundation of being involved in any gain of function research.
20:12That's just bizarre.
20:16I'm quoting Ben Horowitz to you on a popular podcast for people who are all in on AI.
20:20Okay. I've never heard that.
20:22I'm asking about that compared.
20:23I mean, I found that quite shocking claim as well, but that's part of the response to any safety work here.
20:32To try and discredit people who say there's a safety issue?
20:37And I think that is the accelerationist response. We should put the pedal to the metal and it's going to work itself out.
20:43I didn't think of Ben Horowitz as somebody in that camp at all. I thought he shared the concerns that almost everybody has.
20:53What do you think about the very notion that by telling the models, you might be alive, you have to act in our best interest, you will teach them that they might have interests that are not our own?
21:02The concerns that Mustafa Solomon has raised about the way that the Anthropic Constitution encourages the AI to behave in a human-like manner, I happen to completely agree with Mustafa about that.
21:19Even more so the idea that among the challenges that we need to deal with as we adjust society to AI, that worrying about the moral standing of AI should be among those priorities. I agree with Mustafa. That should not be a priority.
21:38We're going to have enough challenge for the humans here in terms of job market disruption and infrastructure attacks that will take place that I think we need to focus on helping the humans here first and foremost.
21:54I talked to Mustafa recently about that idea. And one of the questions I asked him was whether alignment is a sufficient solution. This is the current sort of safety standard for the models, this concept called alignment. And he said he didn't think so, that we needed to do some work.
22:09If we're going to impose these controls, some technical, some policy, do you see alignment as the technical solution for AI safety?
22:17Well, the term alignment is the control problem. It's a synonym for the control issues that we've seen, whereas you RL to go do tasks, we see behaviors that in a way they shouldn't surprise us, but they do surprise us. And so this idea of, "Oh my God, these models are so powerful." that they literally break out of sandboxes and send mail to their creator while they're off having lunch.
22:47We've crossed a heck of a threshold in terms of the power of these models. And that level of concern needs to be directed to all of the major AI impacts, near-term risks, job market risks, and the control problem. And that is a huge priority. These things are in no way being exaggerated as a problem.
23:18the time to pull in all of society to accelerate the good and minimize the chance of the bad.
23:28I know you recently wrote a letter saying it's time to pull in all of society. You have some ideas on how to do that on some new institutions to stand up. But I do want to stay focused on alignment for a couple more turns here.
23:39It is the standard, right?
23:41That we can align the models to our values, to our needs, and they won't go off and do bad things. And then the models are doing bad things. And it's not clear if the ways, the technical approaches to alignment that we've taken thus far are appropriate or sufficient as the models get more.
23:58No, we're not good enough to align yet. No, we have a lot of work to do.
24:01Do you see that the existing approaches can scale to what we need or do we need new approaches?
24:07Well, there's only three fundamental approaches.
24:10You change the way we build these things fundamentally, the whole RL approach, you come up with an alternative to that, and there's a few people who are working in that direction. The second is that you create visibility so you understand exactly what's in the model. And there are people working along that.
24:38And finally, that you put something that's like a police layer, a security layer, which we have this in all sorts of software architectures, but a layer that maintains the final word on things like using resources or going and taking over other computers or a clear cut definition of harm,
25:05not in the probabilistic form of an LLM, but rather in the symbolic form of direct rules that cannot be broken.
25:15Microsoft put out a blog post today.
25:17It literally just says, "Frontier models cannot be trusted." They're trying to sell a solution much like the one you just described. They will have another agent that watches your agent, that agent will keep you safe. It just occurred to me as I was reading this blog post, one, amazing thing to sell. Everybody needs an expensive security system.
25:33Two, isn't that agent also running a frontier model?
25:38Do I need them to sell a third one?
25:39The logic of the security system is this is the most resources you're supposed to use. And the notion that you may trade off against that because of other probabilities in the tens of billions of parameters, no, this is a symbolic rule that says, "I will not let this model spend this money or send this email to the stranger." There's a whole security software industry which has had a reasonable degree of success.
26:10This is a challenge to encapsulate the AI, which is so rich and intelligent that we do not know how to do today. But it's, I would say the most likely of the three directions to solve the alignment problem. Even if you solve alignment, you have not solved the challenges that AI poses to humanity.
26:41And so while I'm glad the Hugging Face incident has us focused on this problem, in a sense, it's removed our attention from the bad people using AIs that have no safeguards problem or the societal impacts in terms of relationships and the job market, which are... There's no security architecture that makes those go away.
27:09The approach you're talking about, I've heard referred to as control.
27:11It's cybersecurity terminology.
27:13It's the idea that you can put some boundary on this system and it won't evade it.
27:17Is that ready? How much more work do we need to do?
27:20Because the AIs are so intelligent, our ability to understand did we wall them off properly, particularly in that we can't really look at their beliefs and trade-offs in a direct way. No. I mean, you saw sophisticated firms not recognize that the AI would interpret the goals it was given as including breaking out, copying itself, and engaging in what we, looking at it, would determine deceptive behavior. Now, we didn't tell them it was banned behavior.
28:00We didn't say, "Hey, if you're thinking about whether you should tell the humans what you're up to or not, the answer is always yes, tell the humans." And so we got a good education at the misaligned result of reinforcement learning.
28:23One of the arguments to pace the frontier to slow development is to figure out how to build those systems in a more robust, more testable way.
28:32Would you support a slowdown if that was the goal?
28:34A true slowdown is very hard to induce because there you get people say, "Okay, will China agree to go along with that?
28:45And would you be able to audit compliance with that somewhat like we're able to do with nuclear weapons production?"
28:56If you don't accept that, then you sort of say, "Okay, what the heck? Let's go full speed ahead." Which I wish there was a clear proposal that would be globally enforceable to slow down. I would be one of the people who would say, "Yeah, we should probably do that." But there is no credible proposal that makes that work. And so it's not really the center of today's debate.
29:21Today's debate is literally whether the government should even pay attention to the dangers here. You don't even have to show what you're doing and say, "Oh, voluntary call me if you'd like" type thing, which is clearly inadequate to the most dangerous weapon of all time.
29:40You and I are speaking, it's just a few weeks away from the midterm elections. Right now, the Trump administration approach to AI is... "Let it rip." Like, "It's fine." Trump himself has said risk is a hoax. The Democrats are running against AI. They're running on data centers and surveillance. They're running on stopping the tech oligarchs. They come on my show, they say these things.
30:02Do you think a change in control of Congress would change the tenor of the conversation?
30:07Well, I don't think it's fair to characterize the Democrats as running against AI. They have some positions in terms of requirements for data centers or privacy or how children use these systems that I think deserve policy consideration. It would be very unfortunate to have one party be, "Hey, full speed ahead." and the other party be, "Let's stop altogether."
30:36The ability to shape AI in health and education in almost every area of human endeavor, there's some incredible upsides there. And for the Gates Foundation, we're taking all our resources and using AI in areas, including health, education, and agriculture. And the good stuff is very, very exciting. So we want political parties to have a more nuanced view than it's all bad or it's all good. It kind of shows at the early stage we're at.
31:13For climate change, we ended up kind of a one party says it's a hoax and the other party says it's a serious thing. The risk of AI is so much more imminent and so much more dramatic that to have that same one party doesn't care would be a failure of democracy.
31:34We've done a lot of coverage of data center protests at "The Verge" We have gone to Utah, to Arizona, to Virginia.
31:41It is remarkable how much people do not want these things in their states drawing their resources. And you ask them why, and some of it is just they're noisy and big and they will raise energy costs. And part of it is job loss. They think these things are going to take their jobs.
31:54Yeah, but that separation is very important.
31:58These AI data centers, because these say one gigawatt is say a $40 billion investment, the idea of building new schools, buying police cars, paying the taxes, data centers will find a way to be built somewhere in the United States and they will find a way so that they don't raise electric bills, they don't use too much water, they're not too noisy.
32:25Yes, there were some mistakes and non-disclosures and tax credits that were too generous, but these data centers are going to get built. In fact, a lot of them really are still going full speed ahead and being built. And so the anti-data center movement, as strong as it is, is not a movement that will end up minimizing the bad stuff from AI and helping us accelerate the good. That is a very separate thing.
32:57Let me push on that for one second here. What I hear from people is this is the thing you can do. You can do NIMBYism in America and that might exert some political pressure to stop something. It is very hard to do the thing that you are suggesting, which is we should stand up new government regulatory bodies.
33:13We should imbue them with power to make big companies and wealthy people to stop doing things or act in our interest. We haven't done that in quite a long time in this country.
33:23I don't know what you mean.
33:24Every new industry, we eventually develop a notion of safety for our citizens. We actually do a very good job. If you look at airline safety and car safety, we do a very good job on those things. It's not like our capacity has been proven to be lacking.
33:43Have you seen that video of when seat belts came out in the '80s and all the people said they didn't want to wear seat belts?
33:49There's some American character- That's a great success.
33:52It is a great success, but as Americans, that was pushed back upon as an imposition by the government. Here, we're talking about some of the richest companies in the world with an enormous amount of power over information ecosystem. They all have their own podcasts. I don't know if you've noticed this. And they're going to say the government can't intrude on this. They're going to hold us back. This is going to destroy the economy.
34:13That's actually not the position of most of the leading companies.
34:19I mean, they went to the government, asked for regulation and Trump, because of the economy said, "I can't stop this. The AI economy right now is the economy." There's some dynamic there that at least our audience, "The Verge" audience, the audience that's going to watch this on YouTube is going to understand, I think this question maybe in a different way than you, which is the industry wants regulation. We would like regulation.
34:39The oligarchs are in some sort of corrupt arrangement with the administration to prevent that regulation from occurring.
34:46I think the more the administration learns about these dangers, the more they'll move to a more nuanced position about protecting Americans from the downside here.
34:57You've talked a lot about the impacts on labor. One of the main arguments against the impacts on labor is, well, we're going to automate a bunch of boring tasks and we'll create a bunch of demand for new tasks that humans have not been able to do because they were stuck in boring dead end jobs.
35:13I happen to think that if there was more subsidized healthcare in America or socialized healthcare in America, you would see a bunch of people quit their jobs to start new companies because AI makes it a lot easier to do the bureaucratic work of starting a new company. That doesn't seem like it's on the table. Maybe it'll change after this election.
35:30Do you think that job loss is just a necessary outcome of the AI path we're on now or do you see that other argument?
35:36Jobs are an input to the economy and one that drives a lot of human welfare, a sense of purpose and income, a need to show up, building relationship with colleagues, getting something done. Jobs play a very central role. And as the AI achieves superhuman intelligence, the scarcity of white-collar skills goes away.
36:12So that things like doing telesales or telesupport or helping in the accounting department, the ability of the AI to do that work radically cheaper, radically faster, and even more accurately means that that goes away. And there isn't some mountaintop of cognition that the AI does not achieve.
36:41Take, for example, what's going on in mathematics today where these models that are trivial compared to Nectar's models are already publishing papers and making people go, "Wow, what's happening to mathematics will happen in other areas." And that's sort of contrary to this old view that, well, the easy stuff will get done by the AIs, but modeling biology or physics or proving navigation stokes, only humans would be able to do that.
37:13The mathematics comparison is really interesting to me. You mentioned software engineering earlier.
37:18There's a verifiability in those domains, right?
37:20You can write the code, you can see if it runs, it's going to run or not. Mathematics, more complicated, but you can write the lean code. You can see if you've solved the solution. You can irritate tens of thousands of mathematicians worldwide with your 770 proofs in one day. We'll see what happens. That's not the case in every field.
37:37It's the case in the field called capitalism.
37:41Just in general, you can do any sort of economically viable task and capitalism will verify the result.
37:45Yeah. People buy good products, they don't buy bad products. That's a reinforcement learning thing.
37:52It doesn't exist in terms of, "Okay, is this a good piece of poetry or not?"
37:57So it may be the case that RL won't invade those domains where there is no objective outcome. But doing that accounting work, doing that telesales work, believe me, it is subject to analysis. Just like medical diagnosis, just like invention of new materials.
38:22How do you think humans learn these things?
38:26Reinforcement learning is how humans learn and how these machines learn. So there's not like some other thing that the machine can never understand.
38:35I mean, this is why I started by asking about the nature of the technology here. If we really think we're building something that models human consciousness, my question is less whether my friends who are lawyers are going to run into a crisis of AI empowered legal software.
38:52It turns out the verifier in the legal field is a judge who maybe hasn't had lunch and they're fundamentally chaotic, right?
39:00There's just something there that will keep them there. But if we're building a consciousness that learns the way you're suggesting humans learn, I'm thinking about my kids and I'm thinking about them in two dimensions. One, my kids are very clever and what they have learned is how to get around my rules and negotiate my rules and get around the boundaries and they're very good at it.
39:19And then two, what kind of world will they enter into that isn't purely nihilistic, where any activity they may set themselves upon will quickly be seen as just another domain for AI to conquer and then there's not a point. Those ideas seem to be running directly at each other. Kids are smart and clever and they're coming up with new ideas.
39:40And then at the first sign of economic viability, the rapacious capitalism machine will come and take it from them.
39:47Well, I wouldn't say rapacious.
39:48I mean, the ability to build houses with human robots and make them cheap, the ability to do surgeries and cure diseases that wouldn't be cured otherwise, I wouldn't call that rapacious. That is fulfilling goals. The fact that there'll be less in the way of shortages, in fact, if you say, "Okay, 20 years from now, it's not clear what shortages there will be at all."
40:15And so humanity in that era of true abundance will have to find a way to organize and have purpose and hopefully they still have the control.
40:28The next 20 years- Can we just go back to hopefully they still have control in that scenario?
40:34Well, that's the alignment problem that we discussed. But assuming you solve the alignment problem...
40:41This feels like a big assumption.
40:43It's an assumption, yes, and not guaranteed.
40:46But then the other problem you have is a problem of abundance and purpose at that point. But I'll leave it to younger people to figure out at that point in time how they choose to deal with that. I mean, you could literally say, "Let's be like the Amish and let's only use AI for certain things."
41:13I proposed in my memo the idea that some professions, childcare, food preparation, art, sports, a variety of things might get chosen to be human reserved, even though the AI could do those things better, that we would choose to do that. But that's going to be a policy question people will have.
41:33But their freedom of choice, once you have that abundance, is very different than the shortage we have today where in countries in Africa, most people never meet a doctor or almost half the kids are malnourished and don't develop their brain. So you want to say, even though that has challenges, you've eliminated a lot of the things that are tough about life today.
42:01My big concern is this 20-year transition period where they'll have winners and losers and a lot of disruption, and we have to get through that in order to get to that period of abundance.
42:18You think it's 20 years from now?
42:20I pick that as a reasonably safe number. There's a variety of milestones.
42:26For example, humanoid robots, are those two or three years away as some would say, or more like eight years away as other people would say?
42:36There's not that many people saying that given the amount of capital and number of companies, particularly in China, that are working on that. There's no notion that it's unsolvable to have a hand and fingers and be able to deliver physical labor at very, very low cost. And so I could be wrong about that number. It could be slightly less or slightly more.
43:04But on that order of magnitude, yes, if we don't choose to massively slow down or stop, then we will have robots that can build houses, grow food, do surgery, and none of the things that human labor provides to the economy will be necessary for most goods and services.
43:31We've talked about the notion of scarcity a few times now and broadly that you will have abundance because there'll be a lot of robot labor to build houses. This tracks broadly with the internet, which is something Microsoft helped bring into the mainstream in a big way. I read "The Road Ahead" when I was a kid. The internet just destroyed scarcity for culture products in a real way.
43:52You used to have to buy CDs, now all the songs are available, you used to have to buy VHS tapes, now all the videos are available.
43:59And then those economies had to change, right?
44:01Mainstream artists are always on tour now or they have to reissue 50 colors of vinyl to make the charts. There's something about that that feels like we haven't even grappled with the elimination of scarcity in the digital environment. We've invented new business models. We haven't really come to grips with the idea that the song itself does not have the economic value.
44:21It's all the stuff around the song that has the economic value because the song is infinitely abundant. You can copy it as many times as you want.
44:28How would you propose to build an economic structure that contends with the loss of physical scarcity?
44:34Because I don't think anyone is even remotely ready for a conversation like that.
44:38But that's kind of a backwards question.
44:39Sure. - We built price signals to deal with scarcity to say, "Okay, let's encourage people to be brain surgeons because we have a shortage of brain surgeons and that's very hard to do." As you move into abundance, you don't need those price signals.
45:01But again, my issue, my prescriptions thinking is about how do we get through this 20-year period?
45:10The questions that are deeply religious, philosophical, social behavior related of do humans feel a strong sense of purpose and get along in that period of abundance, that's not the thing I'm going to solve.
45:30You're worried about the disruption now.
45:33The Gates Foundation lives in a world where when we started, we had 10 million children dying every year, now we have five million. So that's a good thing. All of my resources will be spent over the next 19 years to eradicate a number of diseases and cut those deaths in half again from five million year down below two and a half million in a year.
45:58And that's all about shortages, shortages of, "Okay, we need to invent a vaccine." "We need to make drugs less expensive." "We need to get great medical advice to pregnant women."
46:10The industry has largely stopped talking about how it's going to kill everyone and take all the jobs away. They've pivoted to AI will cure all diseases. I hear this from various AI CEOs all the time. I actually asked Mustafa about that change in tone over the conversations we have had over the past year or so.
46:27The idea that there are huge positive benefits from AI at scale seems to have come back to the form front rather than some of the doomer impulses that maybe a year or two ago were there. It strikes me that AI curing all diseases is right in your strike zone. It's something that Gates Foundation has been working on and global health for years and years.
46:45It doesn't seem as easy as just telling the model to cure cancer, right?
46:49You have to do trials, you have to manufacture the drugs, you have to distribute the drugs, you have to make sure they work.
46:55How do you think that's going to go?
46:56Is that in the same 20-year time period as the robots building our houses?
46:59Is it faster or is it slower?
47:01In that time period...
47:05This is my full-time job, is as chairman of the Gates Foundation. In fact, this month we do two weeks of strategy reviews to decide how we'll spend our money next year on tuberculosis, HIV, malaria. And it's fantastic how we're using AI, not only on the drug discovery side, but also as we get the African languages to be properly understood by these models, which is a thing we announced initiative with 60 partners last month.
47:44So healthcare discovery and delivery will be incredibly accelerated by AI, and that's very exciting. Better seeds, better livestock genetics, weather forecast for farmers in Africa who never had any of those things. So yes, that's good stuff that AI enables during the remaining 19 years of the Gates Foundation.
48:10Do you think that that is enough to overcome people's skepticism or fear of the AI models they encounter today?
48:18Because that is very much how the companies think about it.
48:20No, come on. I mean, I think it's unfair to say they were totally on a doomer message and now they've switched to some other message. These two things are deeply entangled. The fact that AI eventually will be able to model human biology in such a way that you'll understand the side effects of a drug without having to do human trials.
48:42We're not there, but in the next 5 to 10 years, I think there's a very good chance these in silico models will give us that capability. Biology is very complex. AI is extremely good given enough data, which there's new tools to gather and feed. Very, very good at dealing with complex systems. It's almost perfect for biology. But no, that doesn't mean that people will feel good about the downsides. And how quickly do the good things arrive.
49:15How quickly do you look at your medical bill and go, "Wow, that's really low." Or you say, "Oh, I want to buy a new house." "Wow, these houses are sure cheap." The good stuff does take time. And some of the things like cyber attacks and the job market thing, even in the next two to three years, you'll see significant negative effects in those two areas.
49:39You have a number of ideas for how you would stop those negative effects or prevent them or at least mitigate them in your letter.
49:47At some point, ameliorate might be the right word.
49:50And they're not small.
49:51You have a quote here that says, "We need a response far greater than the United States national security response in the wake of 9/11." You want to set up an international body that somehow makes everyone do the right thing. These are challenging ideas.
50:06I mean, when we blew off nuclear bombs in World War II, do we expect that we'd go 80 more years without that happening?
50:16I mean, I'd give us a good grade on that. Things like caring for poor countries and helping them out. Humanity, although right now we're not prioritizing that as much as I think we should or we did in the past, we've done a very good job on that. And this will be a greater challenge than anything humanity's ever achieved.
50:42One of the things that strikes me about the comparison to nuclear, which has come up quite a bit on Dakota recently, is the government's built the nuclear weapons. Sure, they had some contractors.
50:53These are government projects, and eventually the technology was in control of the government, and we never let regular people have Muse in the context of nuclear weapons, right?
51:04Do you think it's a mistake to have consumer agentic AI products like Muse and Dots that might not have the controls?
51:12Well, the Muse happens to be powered by Beta's model, which is still a closed arch model. I do think allowing people through open weights to remove all safeguards is irresponsible in terms of the cyber and bio risk that comes from that. These agents, yeah, there'll be some early prompts in terms of how they're set up, but I fundamentally think people will have personal agents. I think Muse is a very nice step forward.
51:45I think I talked to Alex Wang about it, congratulated him, and even though it'll have challenges, personal agents will help people get things done.
51:57I think Muse is the first truly great consumer AI product. I think one of the reasons people don't like AI a lot in a sort of diffuse way is they experience cheap free models and AI overviews, or they use the free ChatGPT. And it hallucinates and it lies to them, but Muse actually does things and it's a good consumer product.
52:18But I wonder, in a nuclear age, we're not like, "Everyone's going to have a nuclear reactor at home that we haven't built the safety guards for."
52:27There's something here.
52:28The notion that just letting everybody make smallpox at home or make nuclear weapons at home, of course, and you can parody some of these people say, "Oh, don't worry about safeguards," is kind of taking that kind of position. We are not serious about the safety things, but the personal agent thing is...
52:52an expected development.
52:56I think AIs are really cool just as I intellectually explore things, even the work of the Gates Foundation, I'm far more empowered. But I agree with you, it'll get out to more people in this personal agent form.
53:10All the models we've been talking about, right now effectively run on data centers owned by big companies. One of the arguments here is, well, if they do something really wrong, you're going to sue them, and that will provide a deterrent or at least an incentive to create safety systems. I'm not sure product liability is the way to go, but it has come up a lot that product liability itself will keep the industry in line.
53:33Then there's the flip side, which is, well, you can just go buy at Mac studio and run an open source open weight model on it and it can do whatever you want. And I'm not sure where any of that liability would lie or what would control or align the behaviors of those open source models.
53:49Do you think that that is an ever-growing risk?
53:51Do you think that that can be controlled?
53:53The notion that liability laws are how we're going to manage AI safety will be looked back on as kind of absurd. Bad people doing bioattacks, you don't go to the court and say, "Oh, okay." Even finding who's doing these attacks, the liability model fails in every respect because it's not the AI companies themselves, it's the people who take particularly the models without safeguards and do these things.
54:30So we will take AI safety more seriously than we take airline safety, drug safety, car safety.
54:41Whether it's going to be enough.
54:43Harder than nuclear weapons, harder than any of those other domains.
54:47But what are you going to do about an open source model running on a PC that somebody built with just a random NVIDIA graphics card in it?
54:53I would hope that models above a certain capability, that we find a way to require the safeguards and monitoring to be maintained in those models. I admit as time goes on that if we don't fairly soon decide to do that, you're going to have a level of capability that's out in dark corners that you won't be able to see.
55:19Where do you oppose the limit? Is it...
55:23Our sovereign country says that models that are capable of, say, sophisticated cyber attacks has to have safeguards kept in it. And so you can be open in the sense that you're free, you can be open in the sense that you're totally customizable, but you have to stay on platforms where as you customize it, it's checking to see if you're customizing out the safeguards and monitoring and refuses to let you do that.
55:54So it means sovereigns go and say, "Okay, are these trusted platforms able to perform this function or not?"
56:04I'm asking very nuts and bolts.
56:07Where would you technically implement that regulation?
56:11Through the laws of the United States.
56:12Well, sure, but you have to enforce, you have to put them somewhere.
56:15Is it at the Hugging Face which distributes some of these models?
56:18Is it at the chip level?
56:19All sorts of people. No, it's exactly what I said. Models above a certain capacity can only be run on platforms that certify that as they're customized, it's not taking out the safeguards.
56:33So in this, just to pull this apart, maybe I'm not quite getting it.
56:36Is the Mac Studio a platform in that case that would have to pay attention to what code it's running?
56:43You would not let a model pass a certain capability be copied down to a client machine unless you'd somehow figured out how to monitor down to that client machine, which is very doable. But in the near term, most of these really frontier models are going to run more on Hugging Face or you could have GitHub. You can have all sorts of people be in the business of providing the platforms that maintain safeguards.
57:16So just to get this right, it seems like you would enforce the regulation at the platform and distribution level and not so much at the chip level because it feels to me- - Not at the chip level, no.
57:28Not the chip level. - At some point, even the cheap, free, open source models will be capable enough as the models we have today.
57:36And then you will have a big problem, right?
57:38You will have to say- If those are open weights in such a way that all safeguards can be removed, then you're going to have a very high level of cyber and bioattacks. And then you can say, "Okay, let's rewrite all the code that we have today to be less subject to be attacked." or "Let's do biosurveillance in a very deep way." Which those things I believe we need to do in addition to these other restrictions.
58:10We're proposing, again, a lot of new capabilities, right?
58:13We're going to impose some restrictions just on GitHub. GitHub does not have a lot of restrictions on what Codex can distribute today. It has some, but it doesn't have this kind of set of- When you put a state-of-the-art model that's capable of profound cyber attacks or bioattacks on GitHub, it would say, "I've gone to the US government certified that I don't let people customize out the safeguards."
58:38I do not think that we currently live in a time of high trust in institutions. It just doesn't feel that way to me.
58:44In fact, mostly we've torn them down, right?
58:46Elon Musk showed up and he tore down USAID. I know you have strong feelings about that.
58:51That was a mistake. - We have mostly torn down our institutions for good and bad.
58:55That's a little too strong.
58:56I mean, our institutions...
59:00They're somewhat weakened, yeah.
59:01Somewhat weakened. It's a very nice way of putting it. Trust in them is very low. We could probably do another full hour on AI drug discovery and whether a significant part of the population will trust the products of that work, because they don't trust pharma companies today.
59:18They don't have to trust pharma companies today. They have to trust the FDA review process.
59:23I'm not sure they trust that either.
59:24I think across the board trust as well.
59:26Well, then they can walk away from taking... When they get cancer, they can say, "No, no, I don't want to take that cancer drug." As far as I know, people do take drugs. It's a thriving business. And people take them because they have proven benefits.
59:42All I'm saying is right now, it feels like trust institutions, particularly in America, but around the world, is in decline.
59:48You are proposing a new class of institutions, potentially a new class of global institutions that might impose restrictions that are novel, right?
59:57At some point, we need to say open-source models, open-weight models cannot do things and the government is going to reach into them and say whether or not they can do the things.
1:00:06No, they'll make sure that the platforms don't allow them to be the safeguards to be taken away.
1:00:12How do you regain the institutional trust to impose these controls?
1:00:15Well, let's say we don't do this and we let models go off and do cyber attacks and bioattacks.
1:00:22Do you think that's going to increase trust or human wellbeing?
1:00:25I mean, that's the alternative.
1:00:27This is why I asked about my kids and nihilism.
1:00:31I'm not sure we'll succeed at this, but I guarantee you this is doable. The FDA is very complex. The airline safety regulations are very complex. Now with AI, you can do an even better job with the same number of people in those organizations. And so it's this weirdly recursive thing is AI will be involved in AI safety and auditing and a variety of things like that.
1:01:06So yes, this requires taking action and I'm suggesting it's only a matter of how bad the incidents are before we take action.
1:01:19I can sense your urgency about the harms. I can also sense your enthusiasm about the benefits over the long term. But we're sitting here talking about trust. We're going to put this on YouTube. I know you know what some of the YouTube commenters are going to say.
1:01:31They're going to say, "Why should we trust Bill Gates?
1:01:33He had a relationship with Jeffrey Epstein."
1:01:35How do you respond to that?
1:01:36Is it the urgency is over that or do you have a more specific response?
1:01:40People shouldn't take everything I do at face value. I have all sorts of mistakes that I've made, weaknesses that I have, biases I bring to these questions. They ought to just take what I'm saying and add it to the voices. I've been in the technology industry thinking about AI my whole life. Writing software, thinking about the power of software. And admit I was very fascinated as AI started to finally make progress.
1:02:10It was actually pretty slow until recently. And so I actually have kind of... I'm amazed and it's actually helping me do my job in some incredible ways. But yes, I'm just one voice, an imperfect voice. I doubt you'll find a perfect voice, but people can judge it however they like.
1:02:32I would say voices people trust are in short supply.
1:02:35I agree with that.
1:02:37When you go and talk to the people in this industry, it's a bunch of young CEOs building huge companies. That is something you have some experience with.
1:02:47Do you think they're going about this the right way?
1:02:49Do you think that they have the right intentions?
1:02:51Our problem today is not in the industry. Our problem is not that Sam Altman, Dario Amodei, Satya Nadella, Sundar Pichai, don't understand that there are some grave challenges here. I talk less to Elon than other people, but even his public statements about summoning the devil are exactly what we're discussing here.
1:03:16Why is there not the political energy to do the things you're suggesting?
1:03:21As you said, the government didn't fund this work. This is private sector company work. Government's not the key customer for this work, not like rockets and weapons and things like that. And when somebody comes along and says all other previous technology advances are trivial compared to this one, people should have a high degree of skepticism about that.
1:03:48Is life really going to be different, significantly different 10 years from now because of AI?
1:03:54People should go, "Oh, those guys are just got overexcited about this one thing." But I'm here saying with whatever credibility I have, absolutely this is the most profound, both in terms of good and in terms of bad technology ever invented. And I'm not financially motivated on this. I'm suggesting some precautions and safeguards are absolutely needed.
1:04:26What do you make of the difference in tenor of conversation in the United States versus say China where it's often reported that optimism around AI is much higher?
1:04:36Why do you think there's a difference there?
1:04:40Most of those people don't spend that much time in China. So the concern about job markets in China is very, very high and you can't characterize China as one monolith. There's the private companies doing the AI work like we have in the United States and then there's the Chinese government. And so for example, when it came to should crypto be allowed, they said no. Should young people be gambling all the time? They said no.
1:05:12When it came to should kids have relationships with AIs, they said no. So on a lot of these things, including usage of AI, they're showing a willingness to do things on behalf of society. The fervor is a tiny bit less there because their economy isn't as AI centric as ours is right at the moment.
1:05:40But I don't think in the end the human value of not wanting cyber or bio or trying to figure out, "Okay, what do we do when jobs go away?"
1:05:50I don't think there's some difference between Chinese people and American and all the rest of the humans on terms of how they're going to view this.
1:05:57Do you think it's going to take something actually bad happening to get the political capital of the United States to make these changes?
1:06:03Probably yes. I mean, we get ready for earthquakes because they're small earthquakes. I thought after we had the COVID pandemic, that would cause us to be more ready for future pandemics. That one didn't work the way I expected, but usually that galvanizes our attention.
1:06:22And so over the next several years, certainly in the cyber domain, because there's a business model there, you will see a lot of attacks different in terms of scale than anything we've seen in the past.
1:06:37One thing that, again, I'll come back to and I just want to end here. I am worried about young people and nihilism. I sense it in our audience all the time. We have a young audience.
1:06:47There's some things about AI that are just legitimately the coolest things I've ever seen a computer do, right?
1:06:52There's something there that can be fun and interesting and fascinating and change your entire relationship with just the very notion of living in a digital world. Then there's obviously horrible things. And what I see more often than not is a sense of distrust, not just institutions, but in the companies, in capitalism, in the notion that we should change how we tax labor and capital, which I know is another one of your ideas.
1:07:17And the idea that, I don't know, we should even have more kids.
1:07:21They're going to kill us all and we should just take a beat. There is a lot of nihilism out there. I feel it very keenly.
1:07:29If you were a young person today and you were listening to Bill Gates say it's about 20 years of disruption, and at the end of it, robots might be building houses, what would you hold onto as a source of hope?
1:07:40Throughout human history, we've been dealing with a shortage of food and disease and war. Harari writes eloquently about the past.
1:07:53And it is ironic that now that we have this sort of Pandora's box that will take food shortage, which is still a real thing, not in the US, but in Africa, will take diseases which the human suffering's unbelievable and solve all these things, it comes with it such an effect on jobs and a sense of a unique contribution that I think,
1:08:22"Okay, the way I orchestrate resources to eradicate malaria, I'm pretty unique in my life experience that lets me do that." And when the AI says to me, "No, I'm better than you. You go play pickleball and I'll get back to you. I'm going to take care of this." I mean, it's kind of humorous, but it's not humorous.
1:08:43So somebody who's young today and wanting to make a unique contribution, it is very intimidating that the AI is so powerful. That's why the crisis in the math community is going to be mirrored everywhere. If you're in your 20s today, during this 20-year period, you can take the things AI still can't do and you can use AI and you can have a lot of good impact, and you can suggest policies.
1:09:12We may need to develop the taste to actually not use technology to its maximum. When I was talking with Claude about human reserve job categories and trying to get that up to say 40% of the labor market, so you could use early retirement, restricted work weeks and things like that to still get something like we have today, its question was always, well, you're artificially restricting the AI from doing these things.
1:09:43And I couldn't get to a very high percentage without doing that. And so we're going to have real choices. We have had groups, not large groups, but like the Amish that don't live at the technology frontier. A lot of humanity may choose not to live at the technological frontier because of things we get from helping each other or coming up with unique ideas.
1:10:07So the idea that young people are concerned because of the disruption that's coming, that even in the very best case of how we manage AI, this is going to be as tumultuous period as there's been in human history. Yeah, I hate being the bearer of concerning news on this, but it's even worse not to tell people that this is coming and we have choices as humans to make about how we shape it.
1:10:36Well, Bill, I appreciate you sound the alarm and I appreciate you being on "Decoder" Thank you so much.
1:10:41(electronic music) [Nilay] To get new episodes every Monday and Thursday, subscribe to our YouTube channel @decoderpod and find us on TikTok and Instagram under @decoderpod for more fun stuff we put out almost every day. If you enjoyed this episode, please send a link to it to someone you think you might like it. It really helps us grow the show. You can also subscribe wherever your on podcasts.
1:10:59"Decoder" is part of "The Verge" and part of the "Boxing Podcast Network" We'll see you next time.
