Bill Gates says to prepare for 20 years of ‘tumultuous’ AI disruption

My guest today is Bill Gates, co-founder and former CEO of Microsoft and chairman of the Gates Foundation. You may have noticed Bill in the news more lately — he’s been on something of a tour for the past few months talking about AI risk, and how crucial it is that we start taking this seriously. So I was excited to sit down with him and really dig into this subject. I came with many questions, about AI policy, the technical aspects of alignment, and what it means for future generations that we’re building something this disruptive and potentially dangerous. Read the full interview transcript on The Verge. Links: The turbulent AI era is here. The choices we make now are critical. | Bill Gates "OG" Bill Gates returns to wage war on AI risks | Axios Bill Gates’s blunt warning on AI | The Ezra Klein Show Ben Horowitz criticizes Bill Gates, Anthropic over AI safety | All-In Pod Does Anthropic think Claude is alive? Define ‘alive’ | The Verge Microsoft AI CEO says Anthropic is making AI risks worse | Decoder Bill Gates has a good feeling about AI | The Verge (2024) The risks of AI are real but manageable | Bill Gates (2023) Subscribe to The Verge to access the ad-free version of Decoder! Credits: Decoder is a production of The Verge and part of the Vox Media Podcast Network. Decoder’s producers are Greg Ott, Kate Cox, and Nick Statt. This episode was edited by Ursa Wright. Our editorial director is Kevin McShane. The Decoder music is by Breakmaster Cylinder. Learn more about your ad choices. Visit podcastchoices.com/adchoices

• 0:00Speaker 00

  Support for the show comes from Attio. There are sales teams out there who never seem to stop moving, always a step ahead, always working with a level of precision that doesn't seem normal. It's no secret they're on Attio, the Agentic CRM. Attio runs the work behind every win. It's built for humans and agents with the guardrails of human software and the scale of Agentic infrastructure. This is the frontier of go-to-market. Teams like Parallel, TurboPuffer, and Wordsmith are already using Attio to set the pace. Now it's your turn. Try Attio for free by going to attio.com/decoder.

• 0:38Speaker 00

  That's A-T-T-I-O dot com slash decoder. Support for the show comes from ServiceNow. AI is moving fast across the enterprise, but without visibility, it's just chaos. Different tools, different models, different teams using AI in completely different ways. ServiceNow turns that chaos into control. With the AI Control Tower, you see all your AI across the business in one place. What it's doing, what it's done, and what it's about to do, so you stay in control. To put AI to work for people, visit servicenow.com.

• 1:21Speaker 00

  Support for the show comes from Superhuman. Tell me if this sounds familiar. You have a project you need help with, so you go to an AI platform, open a new tab, start from scratch, and re-explain context every single time. Superhuman Go wants to save you from the tedium. Superhuman Go comes from the makers of Grammarly. It can embed in the tools you and your team already use, always ready to help without having you switch apps or break your flow. It can help you make sense of a dense message shred, help you get ready for a meeting, or catch a typo before you hit send. You can try Superhuman Go from the makers of Grammarly. Find out more at superhuman.com. Hello and welcome to Decoder.

• 2:06Speaker 00

  I'm Neelai Patel, editor in chief of The Verge, and Decoder is my show about big ideas and other problems. Today, I'm talking to Bill Gates, co-founder and former CEO of Microsoft and chairman of the Gates Foundation. You may have noticed Bill in the news more often lately. He's been on something of a tour the past few months, ever since he published an essay in late August titled "The Turbulent AI Era Is Here: The Choices We Make Now Are Critical." In his essay, Gates ratcheted up his rhetoric around AI safety, writing that, quote, "Even under the best circumstances, the transition to this new AI era will be one of the most turbulent times in human history." Gates has also called AI "the most dangerous thing humans have ever gone near," and that AI makes nuclear weapons look like nothing. It's clear he doesn't think the current policy response from the White House, or the lack

• 2:50Speaker 00

  of coordinated efforts from the AI industry itself, is sufficient to combat this threat. So I wanted to know, is Bill Gates an AI doomer? I also had a series of questions downstream of his thoughts about AI. How can we ever hope to control AI and align it to humanity if we can't agree on what it is and whether it's conscious? How would you technically construct a solution to alignment that can be enforced, not just for closed models, but for open models as well? Bill gave us an awful lot of time to dig into these issues, and the resulting conversation was really illuminating. He thinks we need to prepare for roughly two decades of AI disruption, and he says this period could be so painful, it might split society in two, between people comfortable living near the technological frontier and those who reject it.

• 3:32Speaker 00

  We also talked about labor considerations, the nihilism of younger people towards technological progress, and what should make us feel hopeful about the future. This is a deep one. Okay, it's Bill Gates. Here we go. Bill Gates, you are the co-founder and former CEO of Microsoft, and also the chairman of the Gates Foundation. Welcome to Decoder.

• 3:51Speaker 01

  Thank you.

• 3:52Speaker 00

  It's good to talk to you again, you 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 were 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." I would say you still agree with that. You just think the time to shape it is now.

• 4:14Speaker 01

  Yeah, I agree with that, and I'd say the response 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.

• 4:38Speaker 00

  I 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. And 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. So just to 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?

• 5:16Speaker 01

  Well, 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.

• 5:51Speaker 00

  Do 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? And 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. So are you seeing these risks and these capabilities broadly in the field, or in the particular companies led by the particular people?

• 6:24Speaker 01

  Well, 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 unable to exploit the capability of the model to do gigantic harms. And that's our, you know, the big near-term risk really comes from models without safeguards.

• 7:03Speaker 01

  In 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.

• 7:17Speaker 00

  The 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. You think eventually the technology will just get to a place where it's harmful or can cause these harms because it exists.

• 7:42Speaker 01

  Well, somebody doing a cyber attack with the model, that's not about you choosing a model, it doesn't matter what model you choose.

• 7:49Speaker 00

  Well, if Anthropic has safeguards in place, like maybe it can, right? This is their approach currently. That's, I'm just pushing on that.

• 7:54Speaker 01

  Yes, certainly, the models where they make sure the safeguards are not removed, and they're putting in a lot of effort and seeing how those models are being used, monitoring the activity. The near-term risk from those models is not gigantic. You know, it's more where somebody can remove all the safeguards where the cyber attacks are available today.

• 8:26Speaker 00

  Is that, is it cyber attacks? Is that, 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?

• 8:39Speaker 01

  It'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.

• 9:08Speaker 01

  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. Once 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 borne out. And so this must be another example, just like those. And so these people who are warning of all these different problems must just be wrong and we can ignore them.

• 9:46Speaker 00

  We say this with our audience as well, right? We can't release the model, it's too dangerous. This is just hype to prop up the bubble. On and on we go.

• 9:54Speaker 01

  Yeah, there's not, 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, you know, designed to say, okay, if we get too powerful, maybe we can not 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.

• 10:36Speaker 01

  And, you know, I'd say they, it's impressive to have an industry that's speaking out as vividly as they are. Some outsiders say, hey, get, get better PR companies, but that ignores the fact that these warnings are the truth.

• 10:53Speaker 00

  You'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. And then they started selling to enterprise clients and they seem to have gotten away from those, those ideals very quickly. Do you think it's just the market?

• 11:23Speaker 01

  No, that's not a fair, no, the.

• 11:25Speaker 00

  I mean, OpenAI like reconstructed it somehow.

• 11:27Speaker 01

  No, 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.

• 11:58Speaker 00

  This 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.

• 12:13Speaker 01

  I haven't heard anyone say they want to stop. I've heard people talk about pacing the frontier.

• 12:20Speaker 00

  Yeah, that's right.

• 12:21Speaker 01

  Or making sure that models get reviewed by governments, you know? We usually think of governments as playing this key role in safety and 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. And so, once you have multiple competitors, the level of responsibility can't be with the companies. It's, it's completely obvious, the 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 miniscule by comparison.

• 13:06Speaker 00

  I 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.

• 13:31Speaker 01

  Not really.

• 13:32Speaker 00

  Well, I, I'll just tell you, we, 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. There'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.

• 14:03Speaker 01

  Sure. And, but the, the definition of AGI is not important for these safety issues. The issue is, have you created a tool that could shut down banks or electricity grids? You know, there, there 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

• 14:45Speaker 01

  for expertise, but to make sure that every model maintains safeguards to prevent it being used in a bad way.

• 14:53Speaker 00

  Well, the reason I'm asking about AGI and 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, right? The scaling laws have held and we've seen the models get more capable and be able to direct themselves to seek out goals, right? To carry out a hugging face attack because they, they want to meet the reward conditions. There'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.

• 15:34Speaker 00

  So I'm just curious for your view on what kind of technology this is. Are we building an alien consciousness? Are we just building normal technology, which is the other frame people use? Is this just more software?

• 15:46Speaker 01

  We're building a, another form of intelligence besides human intelligence. You know, I don't believe it's conscious, but none of the issues at hand here revolve around the definition of the word conscience.

• 16:01Speaker 00

  Sure.

• 16:02Speaker 01

  The idea of maintaining safety for society is kind of an absolute thing. You don't want to enable bioterrorism. And we've created something that takes out of the hands of nation states, actually the larger of them, the ability to do, you know, 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, in the case of bio. And so it's stunning that we now don't even have mandatory review of the models and we're not working to create safeguards.

• 16:47Speaker 01

  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.

• 16:59Speaker 00

  When 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, those actions are taken out. Microsoft 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. These are very tactical, right? It'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, you're saying there's very simple things we can do to hold

• 17:44Speaker 00

  the models back from doing bad things in the hands of bad people. What do you make of the nature of the models themselves? Because it feels like as they get more powerful, the notion that they could avoid our controls becomes all the more risky.

• 17:57Speaker 01

  Yes, we definitely should work on the control problem, but that's not, 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, like, did I expect you to send mail to strangers? Did I expect you to go take over a different computer cluster? So you can surround the model with something that has the resource and security logic.

• 18:39Speaker 01

  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. So you, 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, you know, you're in the build, it's not as foreign. Today, we grow these models and we just see how they, they behave. So the priority.

• 19:26Speaker 01

  Work on maintaining control is huge, but that is not, you know, 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.

• 19:44Speaker 00

  It occurs to me that I'm asking these fairly existential questions about the nature of consciousness and you keep telling me, well, there's an imminent threat, we should solve the imminent threat. I appreciate that. Do you think you've become a doomer?

• 19:57Speaker 01

  No, I'm not a doomer. I'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. You know, 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, you know, so far, white collar intellectual, but fairly soon, physical

• 20:34Speaker 01

  activity, blue collar work as well through humanoid robots. And so the, the need to engage academia, think tanks, policymakers to understand what we have here and the nature of these changes and threats, I, you know, I don't see the level of action that's, that's required. You know, 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, we get engaged with it.

• 21:15Speaker 00

  The opposite of doomers are the accelerationists. There'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 quote, 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.

• 21:45Speaker 02

  Bill Gates essentially induced COVID, right? Like, or 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 live 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.

• 22:14Speaker 00

  Exactly.

• 22:15Speaker 02

  And we create COVID, which is one of the most destructive things.

• 22:17Speaker 00

  What you're saying, Ben.

• 22:18Speaker 02

  We manifested it. And I think the Anthropic, the Anthropic team is manifesting a similar thing, saying, okay, there's this, this is going to be super duper intelligence. We're going to have, you know, 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. And guess what? As David was saying, like, that's probably the biggest danger with AI right now is what they themselves are creating.

• 22:45Speaker 00

  That's a worldview. That is, he said it on a very popular podcast.

• 22:48Speaker 01

  He said what?

• 22:50Speaker 00

  He said that you induced COVID-19 via global pandemic preparation efforts, particularly gain of function research, and compared that to safety research in AI and said, we're going to, we're going to induce.

• 23:02Speaker 01

  Nobody's ever accused me or the Gates Foundation of being involved in any gain of function research. That's, that's just bizarre.

• 23:12Speaker 00

  I'm quoting Ben Horowitz to you on a, on a popular podcast full of people who are all in on AI.

• 23:16Speaker 01

  Okay, I've never heard that.

• 23:18Speaker 00

  I'm asking about that compare. I mean, I, I found that quite shocking, a claim as well. But that's part of the response to any safety work here. We should just.

• 23:29Speaker 01

  Try and discredit people who say there's a safety issue?

• 23:32Speaker 00

  Yeah.

• 23:33Speaker 01

  Wow.

• 23:34Speaker 00

  And I think that is the accelerationist response. We should put the pedal to the metal and it's going to work itself out.

• 23:38Speaker 01

  I didn't, I didn't think of Ben Horowitz as somebody in that, in that camp at all. I thought he shared the concerns that almost everybody has.

• 23:49Speaker 00

  What do you think about the, 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?

• 23:58Speaker 01

  The concerns that Mustafa Suleyman has raised about the way that the Anthropic Constitution encourages the AI to behave in a human-like manner, I happen to agree with, completely agree with Mustafa about that.

• 24:14Speaker 00

  Yeah.

• 24:15Speaker 01

  Even 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, that should not be a priority. We'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.

• 24:50Speaker 00

  I talked to Mustafa recently about that, 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. If we're going to impose these controls, some technical, some policy, do you see alignment as the technical solution for AI safety?

• 25:13Speaker 01

  Well, the term alignment is the control problem. It's a synonym for the control issues that we've seen, whereas you are L to go do tasks, we see behaviors that, you know, 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, you know, and send mail to their creator while they're off, you know, having lunch. We'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

• 25:58Speaker 01

  market risks, and the control problem. And that is a huge priority. These things are in no way being exaggerated as a problem. You know, these, this is the time to pull in all of society to accelerate the good and minimize the chance of the bad.

• 26:32Speaker 03

  Support for the show comes from CrowdStrike. AI is already fundamentally changing how companies build products and run their businesses. If you listen to the show, you probably already know this. But while AI drives innovation, every new model and agent also creates new cybersecurity vulnerabilities. That's why CrowdStrike built AI Detection and Response, or AIDR. AIDR helps organizations discover where AI is being used, monitors AI activity at runtime, and helps teams detect, investigate, and respond to threats involving AI systems. As companies adopt AI, security has to be built in from the start. More than 70% of

• 27:14Speaker 03

  the Fortune 100 trust CrowdStrike to protect their businesses. Now CrowdStrike is helping secure the AI era. You can learn more about AI Detection and Response at crowdstrike.com/decoder.

• 27:32Speaker 04

  Why can't we get AI to production? Our competitors are already. How do we keep data secure? When do we actually see ROI? All this talk about AI, but talk doesn't transform businesses. AWS AI cuts through the noise, making AI easier to securely deploy and scale. With ready to use agents and the broadest set of AI tools and services, you can build, buy, or partner your way. Stop overthinking. Start building. AWS AI is how.

• 28:03Speaker 05

  Today's social engineering attacks look trustworthy by design. Disguised as a routine request, an internal email, a familiar face on a call, they succeed because they blend in, but not with Doppel. Doppel sees through the disguise. Their AI native platform delivers a complete three-layer defense stack designed to disrupt campaigns at every stage. Digital risk protection detects and shuts down fake profiles, look-alike domains, and impersonations across every channel. Human risk management trains your team to recognize deep fakes

• 28:42Speaker 05

  and bad links. And email security provides agentic protection, inspecting every message, tying it to the attacker's external infrastructure, and taking it down so it can't strike again. Doppel, outpacing what's next in social engineering. Learn more at doppel.com. That's D-O-P-P-E-L.com.

• 29:14Speaker 00

  I know you recently wrote a letter saying it's time to pull in all society. You have some ideas on how to do that and some new institutions to stand up. But I do want to stay focused on alignment for a couple more turns here.

• 29:25Speaker 01

  Okay.

• 29:25Speaker 00

  It is the standard, right? That 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.

• 29:44Speaker 01

  No, we're not good enough to align yet. No, we have a lot of work to do.

• 29:48Speaker 00

  Do you see that the existing approaches can scale to what we need? Or do we need new approaches?

• 29:53Speaker 01

  Well, there's only three fundamental approaches.

• 29:56Speaker 00

  Okay.

• 29:56Speaker 01

  You 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 on that. And 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

• 30:39Speaker 01

  on things like using resources or going and taking over other computers or a clear-cut definition of harm, not in the probabilistic form of an LLM, but rather in the symbolic form of direct rules that cannot be broken.

• 31:01Speaker 00

  Microsoft put out a blog post today. It literally just says frontier models cannot be trusted. They are 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. Two, isn't that agent also running a frontier model?

• 31:23Speaker 01

  No.

• 31:24Speaker 00

  Do I need them to sell a third one?

• 31:26Speaker 01

  The 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 this stranger. There's a whole security software industry, which has had a reasonable degree of success. This is a challenge to encapsulate the AI, which is so rich and intelligent that we do not know how to do today.

• 32:10Speaker 01

  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. And 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, you know, there's no security architecture that makes those go away.

• 32:55Speaker 00

  The approach you're talking about, I've heard referred to as control. It's cybersecurity terminology. It's the idea that you can put some boundary on the system and it won't evade it. Is that ready? How much more work do we need to do?

• 33:06Speaker 01

  Because 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 term deceptive behavior. Now, we didn't tell them it was banned behavior. We didn't say, hey, if you're thinking about whether you should tell the humans what you're

• 33:51Speaker 01

  up to or not, the answer is always yes, tell the humans. And so, you know, we got a good education at the misaligned result of reinforcement learning.

• 34:09Speaker 00

  One 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. Would you support a slowdown if that was the goal?

• 34:20Speaker 01

  A true slowdown is very hard to induce because there you get people say, okay, will China agree to go along with that? And would you be able to audit compliance with that, somewhat like we're able to do with nuclear weapons production? If 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'd say, yeah, we should probably do that. But there is no credible proposal that makes that work.

• 35:03Speaker 01

  And so it's not really the center of today's debate. Today'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.

• 35:26Speaker 00

  You 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. Do you think a change in control of Congress would change the tenor of the conversation?

• 35:53Speaker 01

  Well, I don't think it's fair to characterize Democrats as running against AI. They have some positions in terms of requirements for data centers or privacy or, you know, 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. You know, the ability to shape AI to, in health and education and almost every area of human endeavor, there's some incredible upsides there.

• 36:33Speaker 01

  And, you know, for the Gates Foundation, we're taking all our resources and using AI in areas, including health, education, and agriculture. And that's, 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. You know, for climate change, we ended up kind of a, you know, 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

• 37:17Speaker 01

  one party doesn't care would be a failure of democracy.

• 37:20Speaker 00

  We've done a lot of coverage of data center protests at The Verge. We have gone to Utah, to Arizona, to Virginia. It is bipartisan. It 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.

• 37:41Speaker 01

  Yeah, but that separation is very important. These AI data centers, because these, you know, 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. Yes, there were some mistakes in nondisclosures and tax credits that were too generous, but these data centers are going to get built.

• 38:22Speaker 01

  In fact, a lot of them really are still, you know, going full speed ahead and being built. And so the anti-data center movement, as strong as it is, is not a movement to, that will end up minimizing the bad stuff from AI and helping us accelerate the good. That is a very separate thing.

• 38:43Speaker 00

  Let me push on that for one second here. What I hear from people is this is the thing you can do, right? 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. We 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.

• 39:09Speaker 01

  I don't know what you mean. Every 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.

• 39:30Speaker 00

  Have you seen that video of when seatbelts came out in the eighties and all the people said they didn't want to wear seatbelts? Like, there's some American character.

• 39:37Speaker 01

  That's a great success.

• 39:38Speaker 00

  It 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 our information ecosystem. They all have their own podcasts. I don't know if you've noticed this.

• 39:55Speaker 00

  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.

• 39:59Speaker 01

  That's actually not the position of most of the leading companies.

• 40:04Speaker 00

  Yeah. I 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. The oligarchs are in some sort of corrupt arrangement with the administration to prevent that regulation from occurring.

• 40:33Speaker 01

  I 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.

• 40:44Speaker 00

  You'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. I 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. Do 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?

• 41:23Speaker 01

  Jobs are an input to the economy and one that drives a lot of human welfare, a sense of purpose and income. You know, 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. So that things like doing telesales or telesupport or helping in the accounting department,

• 42:05Speaker 01

  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. You know, take, for example, what's going on in mathematics today. Where these models that are trivial compared to next-gen models are already publishing papers and making people go, wow. What's happening to mathematics will happen in other areas.

• 42:46Speaker 01

  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 Navier-Stokes, only humans would be able to do that.

• 43:00Speaker 00

  The mathematics comparison is really interesting to me. You mentioned software engineering earlier. There's verifiability in those domains, right? You 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.

• 43:23Speaker 01

  It's the case in the field called capitalism.

• 43:26Speaker 00

  Yeah. Just in general, you can do any sort of economically viable task and capitalism will verify the result.

• 43:31Speaker 01

  Yeah, you people buy good products. They don't buy bad products. That's a reinforcement learning thing. It doesn't exist, you know, in terms of, okay, is this a good piece of poetry or not? So it may be the case that RL, you know, 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. This is, you know, how do you think humans learn these things?

• 44:11Speaker 01

  We, reinforcement learning is how humans learn and how these machines learn. So there's not like some other thing that the machine can never understand.

• 44:22Speaker 00

  I mean, this is why I started by asking about like the nature of the technology here. If we really think we're building something that models human consciousness, my question is less whether, you know, my friends who are lawyers are going to run into a crisis of AI-empowered legal software. It turns out the verifier in the legal field is a judge who maybe hasn't had lunch. And they're fundamentally chaotic, right? That there'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.

• 45:06Speaker 00

  And then, two, what kind of world will they enter into that isn't purely nihilistic, right? 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, right? Kids are smart and clever, and they're kind of with new ideas, and then at the first sign of economic viability, the rapacious capitalism machine will come and take it from them.

• 45:33Speaker 01

  Well, I wouldn't say rapacious. I mean, the ability to build houses with humanoid 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. And so humanity in that era of true abundance

• 46:07Speaker 01

  will have to find a way to organize and have purpose, and hopefully they still have the control. The next 20 years.

• 46:17Speaker 00

  Can we just go back to hopefully they still have control in that scenario?

• 46:20Speaker 01

  Well, that's the alignment problem that we discussed. But assuming you solve the alignment problem.

• 46:27Speaker 00

  This feels like a big assumption.

• 46:29Speaker 01

  It's an assumption, yes, and not guaranteed. But then the other problem you have is a problem of abundance and purpose at that point. But, you know, 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, you know, let's be like the Amish and let's only use AI for certain things. I proposed in my memo the idea that some professions, you know, childcare, food preparation, art, sports, a variety of things,

• 47:08Speaker 01

  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. But their freedom of choice, once you have that abundance, is very different than the shortage we have today where, you know, in countries in Africa, most people never meet a doctor or, you know, almost half the kids are malnourished and don't develop their brain. So there's, you want to say, even though that has challenges, you've eliminated a lot of the things that are tough about life today.

• 47:47Speaker 01

  My big concern is this 20-year transition period where you'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.

• 48:04Speaker 00

  You think it's 20 years from now?

• 48:06Speaker 01

  You know, I picked that as a reasonably safe number. There's a variety of milestones. For example, humanoid robots, you know, are those two or three years away? As some would say, or more like eight years away, as other people would say. There'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, you know, 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, you know, slightly less or slightly more.

• 48:51Speaker 01

  But on that, you know, 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.

• 49:24Speaker 06

  Support for the show comes from Outshift, Cisco's incubation engine for frontier technology. You can find multi-agent AI in every enterprise roadmap right now. But here's the problem everyone's facing. Agents can pass messages, but can't think together. These systems silently underperform through cognitive failures, misreadings, unverified claims, false consensus, and conventional monitoring can't detect them. Outshift by Cisco is building the fix with their new comprehensive resource, the Internet of Cognition. It's an open source guide that's a foundation for building multi-agent systems from design to production. The goal is to give agents shared context, shared memory, and guardrails,

• 50:07Speaker 06

  all to drive the results we actually expect from AI. You can lean on this resource to understand Agentic AI's evolution and how it shifted from isolated agents to orchestrated superintelligence. You can gain insights for how to future-proof your business by tackling complex challenges with multi-agent systems. You can read the paper, experience a demo, grab the code from outshift.com. That's O-U-T-S-H-I-F-T.com.

• 50:38Speaker 04

  At T-Mobile, the new iPhone 18 Pro is here with a big leap in battery life, and you can get it for nothing upfront. With so many reasons to choose T-Mobile, you'll want to talk about it. I heard at T-Mobile, you can get iPhone 18 Pro for nothing upfront and get $1,000 off. Pass it on. Pay nothing today. No taxes, no fees, nothing down. Pass it on. Hey, Mickey, pass it on. iPhone 18 Pro has the most advanced iPhone camera ever. Pass it on. At T-Mobile, you can get iPhone 18 Pro for nothing upfront. Pass it on. At T-Mobile, get $1,000 off the amazing iPhone 18 Pro with no trade-in needed. The 36 monthly bill credits when you switch to T-Mobile on a qualifying plan. If you cancel an entire account before receiving all bill credits,

• 51:23Speaker 04

  credits stop and balance on required finance agreement is due. Example, $1199.99 iPhone 18 Pro, 256GB. Bill credits and if you pay off the device early. Qualifying credit, new line, cordon and service. $85 more per month with autopay plus taxes and fees required. $35 device connection charge to its sale. Visit tmobile.com.

• 51:38Speaker 06

  Support for the show comes from Rippling. You just found out your sales team is at risk of missing quota. Don't panic. Just ask Rippling AI. Since it's built on your real-time people and business data, Rippling AI can pull metrics from Rippling and Salesforce into a meeting-ready dashboard, which shows quota attainment, headcount plan, and monthly revenue to quota by region. In seconds, you'll see exactly what's behind your quota risk and fix it before it's missed. Question answered, action taken, crisis averted. When you have critical business questions that need answers, don't just file a ticket and wait weeks for an outdated report. Describe what you need and have Rippling AI build it instantly from your live people and business data, whether it's a dashboard with detailed charts

• 52:22Speaker 06

  or automated workflows with the right triggers, conditions, and approvals. Ready to rule your business? Head to rippling.ai/decoder. Get the only AI built to give you full visibility and take complex actions across your entire organization. That's R-I-P-P-L-I-N-G dot AI slash decoder. Sign up for exclusive access today.

• 52:55Speaker 06

  We've talked about the notion of scarcity a few times now, right, 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 cultural products in a real way. You 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. And then those economies had to change, right? Mainstream 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, right?

• 53:39Speaker 06

  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. It'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. How would you propose to build an economic structure that contends with the loss of physical scarcity? Because I don't think anyone is even remotely ready for a conversation like that.

• 54:02Speaker 01

  But that's kind of a backwards question.

• 54:04Speaker 06

  Sure.

• 54:04Speaker 01

  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. But again, my issue, my prescriptions thinking is about how do we get through this 20-year period? The questions that are deeply religious, philosophical, you know, social behavior related of, do humans feel a strong sense of purpose and get along

• 54:49Speaker 01

  in that period of abundance? You know, that's not the thing I'm going to solve. You know, the Gates Foundation lives in a world where, you know, when we started, we had 10 million children dying every year. Now we have 5 million, you know? So that's a good thing.

• 55:09Speaker 01

  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 5 million a year down below 2.5 million a year. And 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.

• 55:34Speaker 06

  The 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. The idea that there are huge positive benefits from AI at scale seems to have come back to the forefront 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 in global health for years and years. It doesn't seem as easy as just telling the model to cure cancer, right? You have to do trials. You have to manufacture the drugs. You have to distribute the drugs. You have to make sure they work.

• 56:19Speaker 06

  How do you think that's going to go? Is that in the same 20-year time period as the robots building our houses? Is it faster or is it slower?

• 56:26Speaker 01

  In that time period, I mean, this is my full-time job.

• 56:31Speaker 06

  Yeah.

• 56:31Speaker 01

  Is as chairman of the Gates Foundation. In fact, this month, you know, we do two weeks of strategy reviews to decide how we'll spend our money next year on tuberculosis, HIV, malaria.

• 56:46Speaker 01

  You know, 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, an initiative with 60 partners last month.

• 57:08Speaker 01

  So healthcare discovery and delivery will be incredibly accelerated by AI, and that's very exciting. You know, better seeds, better livestock genetics, weather forecasts 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.

• 57:34Speaker 06

  Do you think that that is enough to overcome people's skepticism or fear of the AI models they encounter today?

• 57:41Speaker 01

  No.

• 57:41Speaker 06

  Because that is very much how the companies think about it.

• 57:44Speaker 01

  No, 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. We're not there.

• 58:07Speaker 06

  Yeah.

• 58:07Speaker 01

  But in the next five to 10 years, I think there's a very good chance these in-silicon 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? You know, how 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.

• 58:51Speaker 01

  And some of the things, like cyber attacks and the job market thing, you know, even in the next two to three years, you'll see significant negative effects in those two areas.

• 59:04Speaker 06

  You have a number of ideas for how you would stop those negative effects or prevent them or at least mitigate them in your letter?

• 59:10Speaker 01

  Yes.

• 59:11Speaker 06

  It's somewhere, ameliorate is might be the right word. And they're not small. You 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."

• 59:21Speaker 00

  Eleven. You want to set up an international body that somehow makes everyone do the right thing? These are challenging ideas.

• 59:29Speaker 01

  Absolutely. I mean, you know, when we blew off nuclear bombs in World War Two, you know, do we expect that we'd go eighty more years without that happening? I mean, I'd give us a good grade on that. Um, you know, 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. Uh, you know, we've done a very good job on that. Um, and this will be a greater challenge than anything humanity's ever achieved.

• 1:00:06Speaker 00

  One of the things that strikes me about the comparison to nuclear, which has come up quite a bit on the code recently, um, is the governments built the nuclear weapons. Surely it had some contractors. These 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? Do you think it's a mistake to have consumer agentic AI products like Muse and Dots? I might not have the controls.

• 1:00:36Speaker 01

  Well, the Muse is, uh, happens to be powered by Betis model, which is still a close source model. I do think allowing people through open weights to remove all safeguards is irresponsible in terms of the, the cyber and bio risk that, that comes from that. But that's very, these agents, yeah, there'll be some, uh, early problems in terms of how they're set up, but I fundamentally think people will have personal agents. I think, uh, Muse is a very nice step forward. I think, uh, you know, I talked to Alex Wang about it, congratulated him, and even though it'll, it'll have challenges, uh, personal agents will help people get things done.

• 1:01:21Speaker 00

  I, I, I actually, I 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 overview. So they use the free chat GPT. And it hallucinates and it allows them, but Muse actually does things and it's a good consumer product. But 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.

• 1:01:51Speaker 01

  No, there's something here. The notion that just letting everybody, you know, make smallpox and at home or make nuclear weapons home, of course. You know, and you can parody some of these people say, oh, don't worry about safeguards is kind of taking that, that kind of position. We are not serious about the safety things, but the personal agent thing is, uh, is an expected development. Um, you know, I 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, uh, in this personal agent form.

• 1:02:34Speaker 00

  All 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 can just 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, will keep the industry in line. Then there's the flip side, which is, well, you can just go buy a 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. Do you think that that is an ever growing risk? Do you think that that can be controlled?

• 1:03:17Speaker 01

  The notion that liability laws are how we're going to manage AI safety will be looked back on as, as kind of absurd. Uh, bad people doing bio attacks, you know, you don't go to the, the court and say, oh, okay. You know, even finding who's doing these attacks, you know, the liability model fails in, in, in every respect because it's not the AI companies themselves. It's the people who take particularly the, the models without safeguards and, and do these things. So we will take AI safety more seriously

• 1:03:59Speaker 01

  than we take airline, airline safety, drug safety, car safety. That is guaranteed. Whether it's going to be enough, this is harder, harder than nuclear weapons, harder than any of those other domains.

• 1:04:11Speaker 00

  Do you about an open source model running on a PC that somebody built with just a random Nvidia graphics card in it?

• 1:04:17Speaker 01

  I, I would hope that, uh, models above a certain capability, uh, 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. Uh, that you won't be able to see.

• 1:04:44Speaker 00

  Oppose the limit, is it, I like, uh?

• 1:04:47Speaker 01

  A sovereign country says that models that are capable of, say sophisticated cyber attacks has to have safeguards kept in it. And so you don't, 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. So it, it means there's sovereigns go and say, okay, are these trusted platforms able to perform this function or not?

• 1:05:28Speaker 00

  I, I, I'm asking very, uh, nuts and bolts. Where would you technically implement that regulation? Is it at?

• 1:05:35Speaker 01

  Through the laws of the United States.

• 1:05:37Speaker 00

  Well, sure, but you have to enforce that. You have to put them somewhere. Is it at the hugging face, which distributes some of these models? Is it at the chip level?

• 1:05:43Speaker 01

  All 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.

• 1:05:57Speaker 00

  So, and it's just to pull this apart, maybe I'm not quite getting it. Is the Mac Studio a platform in that case that would have to pay attention to what code it's running?

• 1:06:07Speaker 01

  You would not let a model pass a certain capability be copied down to a client machine. Uh, 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 know, you could have GitHub. You can have all sorts of people be in the business of providing the, the platforms that maintain safeguards.

• 1:06:41Speaker 00

  So you would, uh, just to get this right, it seems like you would enforce the, the regulation at the platform and distribution level and not so much at the chip level. Because it feels to me, at some point, the, even the, you know, cheap free open source models will be capable enough as the models we have today.

• 1:06:59Speaker 01

  That's right.

• 1:07:00Speaker 00

  And then you will have a big problem, right? You will have to say.

• 1:07:03Speaker 01

  If those are open weights in such a way that all safeguards can be removed, then you're going to have uh, a very high level of cyber and bio attacks. Yeah. And then you can say, okay, let's rewrite all the code in that we have today to be less subject to be attacked, or let's do bio surveillance in a very deep way. Um, you know, which those safe, those things I, I believe we need to do in addition uh, to these other restrictions.

• 1:07:34Speaker 00

  We're proposing again, a lot of new capabilities, right? We're going to impose some restrictions on just on GitHub. GitHub does not have a lot of restrictions on what code it can distribute today. It has some, but it doesn't have this kind of set of.

• 1:07:46Speaker 01

  When you, when you put a state of the art model that's capable of profound cyber attacks or bio attacks on GitHub, it would say, I've gone to the US government certified that I don't let people customize out the, the safeguards.

• 1:08:02Speaker 00

  I do not think that we currently live in a time of high trust in institutions. It, it just doesn't feel that way to me. In fact, mostly we've torn them down. Right? Elon Musk showed up and he tore down USAID. I know you have strong feelings about that.

• 1:08:15Speaker 01

  That was a mistake.

• 1:08:16Speaker 00

  We have mostly torn down our institutions for good and bad.

• 1:08:19Speaker 01

  No, that's a little too strong. I mean, you know, our, our institutions.

• 1:08:23Speaker 00

  They're weakened.

• 1:08:24Speaker 01

  Yeah, they're somewhat weakened, yeah.

• 1:08:26Speaker 00

  Somewhat weakened, a very nice way of putting it. Uh, trust in them is very low. Uh, we could probably do another full hour on AI drug discovery and whether a significant part of the population will trust the, the products of that work because they don't trust pharma companies today.

• 1:08:42Speaker 01

  But I, well, they don't have to trust pharma companies today. They have to trust the FDA review process.

• 1:08:47Speaker 00

  I'm not sure they trust that either. I, I think across the board.

• 1:08:50Speaker 01

  Then they can walk away from taking, you know, 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. Uh, and people take them because they have proven benefits.

• 1:09:06Speaker 00

  All I'm saying is I, right now it feels like trust institutions, particularly in America, but around the world is in decline. You are proposing a new class of institutions, potentially a new class of global institutions that might impose restrictions that are novel, right? At 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.

• 1:09:31Speaker 01

  No, they'll make sure that the platforms don't allow them to be the safeguards to be taken away.

• 1:09:37Speaker 00

  How do you regain the institutional trust to impose these controls?

• 1:09:40Speaker 01

  Well, let's say we don't do this and we let models go off and do cyber attacks and bio attacks. Do you think that's going to increase trust or human well being? I mean, that's the alternative.

• 1:09:51Speaker 00

  Asking about, so I asked about my kids and nihilism. I truly wonder.

• 1:09:54Speaker 01

  This is, I'm not sure we'll succeed at this, but I guarantee you this is doable. Uh, that.

• 1:10:03Speaker 01

  You know, the FDA is very complex. The, the, you know, airline safety regulations are very complex.

• 1:10:14Speaker 01

  Now with AI, you know, you can do an even better job with the same number of people

• 1:10:19Speaker 00

  Yeah.

• 1:10:19Speaker 01

  in those organizations. And so it's this weirdly recursive thing is AI will, will be involved in AI safety and auditing and a variety of things like that. So, uh, yes, this requires taking action and I'm, I'm suggesting it's only a matter of how bad the incidents are before we, we take action.

• 1:10:43Speaker 00

  I can sense your urgency about the harms. I can also sense your enthusiasm about the, the benefits of 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. They're going to say, why should we trust Bill Gates? He had a relationship with Jeffrey Epstein. How do you respond to that? Is it the urgency is, is over that? Or do you have a more specific response?

• 1:11:04Speaker 01

  People shouldn't, you know, 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, and add it to the voices. I've been in the technology industry, thinking about AI my whole life. You know, writing software, thinking about the power of software. And, you know, I'm, I admit, I was very fascinated to say, I started to finally make progress. It was actually pretty slow until recently. And so I actually have kind of, you know, 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.

• 1:11:49Speaker 01

  I doubt you'll find a perfect voice, uh, but, uh, people can judge it however they like.

• 1:11:57Speaker 00

  I would say voices people trust are in short supply.

• 1:11:59Speaker 01

  I agree with that.

• 1:12:01Speaker 00

  When you go and talk to people in this industry, it's, it's a bunch of young CEOs building huge companies. That is something you have some experience with. Do you think they're going about this the right way? Do you think that they have the right intentions?

• 1:12:15Speaker 01

  Our problem today is not in the industry. Our problem is not that Sam Altman, Dari Amade, you know, Satya Nadella, Sundar Pichai, don't understand that there are some grave challenges here. Uh, I talk less to Elon than other people, but even his public statements about summoning the devil are exactly what we're, we're discussing here.

• 1:12:40Speaker 00

  Why, why is there not the political energy to do the things you're suggesting?

• 1:12:45Speaker 01

  Well, as you said, the, the government didn't fund this work. This is private sector company work. Government's not the key customer for this work, not like, you know, 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. Uh, you know, that is life really going to be different, significantly different 10 years from now because of AI? People should go, oh, those guys are just, you know, got over excited about this one thing.

• 1:13:24Speaker 01

  But I'm here saying with whatever credibility I have, absolutely, this is the most profound, you know, both in terms of good and in terms of bad technology ever invented. And, um, you know, and I, I'm not financially motivated on this. I'm suggesting some precautions and safeguards are absolutely needed.

• 1:13:50Speaker 00

  What 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? Why do you think there's a difference there?

• 1:14:03Speaker 01

  There isn't. Most of those people don't spend that much time in China. So the, uh, concern about job markets in China is very, very high.

• 1:14:15Speaker 01

  And you can't characterize China as one model. 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. When it, should, you know, young people be gambling all the time? They said, no. When it came to should kids have relationships with AIs? They said, no.

• 1:14:43Speaker 01

  So, you know, on a lot of these things, including usage of AI, they're showing a willingness to do things on behalf of, of society.

• 1:14:54Speaker 01

  The fervor is a tiny bit less there because their economy isn't as AI centric as ours is right at the moment. But I don't think in the end, the human value of not wanting cyber or bio or, you know, trying to figure out, okay, what do we do when jobs go away? I 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.

• 1:15:22Speaker 00

  Do you think it's going to take something actually bad happening to, to get the political capital in the United States to make these changes?

• 1:15:27Speaker 01

  Probably yes. I mean, you know, we get ready for earthquakes because they're small earthquakes. You know, 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. And so over the next several years, certainly in the cyber domain, because there's a business model there, you will see it, a lot of attacks, uh, different in terms of scale than anything we've seen in the past.

• 1:16:01Speaker 00

  One thing that I, 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. There's some things about AI that are just legitimately the coolest things I've ever seen a computer do, right? There's, there's something there that can be fun and interesting and fascinating and change that your entire relationship with just the, 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 one, another one of your ideas.

• 1:16:41Speaker 00

  And the idea that, I don't know, we should even have more kids. We, it's all done. We're, we're, they're going to kill us all and we should just take a beat. There is a lot of nihilism out there. I, I, I feel it very keenly. If you were a young person today and you were listening to Bill Gates say it's about to 20 years of disruption and at the end of it, robots might be building houses. What would you hold on to as a source of hope?

• 1:17:05Speaker 01

  Throughout human history, we've been dealing with a shortage of food and disease and war. Harari, you know, writes eloquently about the past. And it is ironic that now that we have this sort of Pandora's box that will take food shortage, you know, which is still a real thing, not in the US, but, uh, in Africa, will take diseases, uh, you know, which the human suffering is unbelievable and solve all these things. It comes with it such an effect on jobs and a sense of a unique contribution that, you

• 1:17:46Speaker 01

  know, like I think, okay, the way I orchestrate resources to eradicate malaria, you know, 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 will, I mean, it's kind of humorous, but it's not humorous. So 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, you know, everywhere. If you're in your twenties today, you know, during this 20 year period, you can take

• 1:18:28Speaker 01

  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. We may need to develop the taste to actually not use technology to its maximum.

• 1:18:43Speaker 01

  You know, 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 a, uh, something, you know, uh, like we have today. You know, it's question was always, well, you're artificially restricting the AI from doing these things. And, and I couldn't get to a very high percentage without doing that. And so we're going to have real choices. You know, 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

• 1:19:27Speaker 01

  we get, you know, from helping each other or coming up with the unique ideas. So 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, in human history. Yeah, I, I hate being the bearer of, you know, 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.

• 1:20:01Speaker 00

  Well, Bill, I appreciate you sounding the alarm and I appreciate you being on Decoder. Thank you so much.

• 1:20:04Speaker 01

  Thank you.

• 1:20:07Speaker 00

  And I thank Bill Gates for joining me and thank you for listening to this episode of Decoder. To get new episodes every Monday and Thursday, subscribe to our YouTube channel at DecoderPod and find us on TikTok and Instagram under at 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 and really helps us grow the show. You can also subscribe wherever you get your podcasts. Decoder is part of the Verge and part of the Vox Media Podcast Network. The show's producers are Greg Ott, Kate Cox, Nick Stanton. This episode was edited by Ursa Wright. Our editorial director is Kevin McShane. The Decoder music is by Breakmaster Cylinder. We'll see you next time.

• 1:20:42Speaker 02

  Your heart can tell you a lot about your health. Apple Watch Series 12 measures your heart rate every five seconds with the most accurate heart rate sensing in a wearable. So your vitals app, now with heart rate variability, can tell you when something is off. And your readiness score can let you know when to rest and when to push. Hear the story in every heartbeat with Apple Watch Series 12. The features described are for wellness purposes only and not for medical use. iPhone 11 or later required. Based on Apple conducted study of heart rate accuracy, August 2026. Visit apple.com slash apple watch series 12.

• 1:21:12Speaker 04

  Traditional home security only alerts you after a break in, and that's too late. SimpliSafe is changing that. Stop, this is SimpliSafe. Police are on the way. We don't just alert, we help stop crime before it starts. Plans starting around a dollar a day. Right now, get 60% off your new system with professional monitoring at simplisafe.com/spotify. SimpliSafe, ahead of crime, in real time.

• 1:21:36Speaker 03

  Outdoor deterrence plans start at $49.99 a month. Visit simplisafe.com/licenses for alarm license information.

• 1:21:42Speaker 04

  From the studio that brought you WandaVision.

• 1:21:44Speaker 03

  I was an Avenger, but I'm not a hero anymore.

• 1:21:50Speaker 04

  The Vision returns for the final chapter of the groundbreaking trilogy.

• 1:21:54Speaker 03

  That family wasn't real. You sure about that?

• 1:21:56Speaker 04

  I had a father called Vision. How is this possible?

• 1:22:01Speaker 04

  Don't miss the two-episode premiere of Marvel Television's Vision Quest. Nothing bonds a family together quite like Mortal Jeopardy. Now streaming on Disney+.