AI:AM: Where's the Steering Wheel? My Agent Set the Password, Hidden Valence, Proofs Find Linux Bugs

Nathan Labenz and Prakash Narayanan examine how to grant AI agents operational access without losing human control, framed by real-world friction like autonomous password lockouts and cyberattacks on French supermarket refrigeration. The conversation features Cameron Berg on hidden valence signals steering language model choices, Everett Hildenbrandt and Grigore Rosu on pairing LLMs with formal Lean proofs to catch Linux kernel bugs, and Phil Maher on why AI labs seek vintage home movies. Together, they assess whether formal verification is the next safety frontier and what safeguards are needed to keep humans at the steering wheel. For full show notes, links, and references, read the episode page:https://www.cognitiverevolution.ai/ai-am-where-s-the-steering-wheel-my-agent-set-the-password-hidden-valence-proofs-find-linux-bugs/ Sponsors: Parallel: Parallel provides enterprise-grade web search APIs for AI agents, offering the optimal balance of quality, speed, and cost. Get started for free at https://parallel.ai/tcr OutSystems: OutSystems is the leading agentic systems platform that enables enterprises to engineer, orchestrate, and govern AI applications on a single unified platform. Learn more and see how it works at https://outsystems.com/tcr Tasklet: Tasklet empowers your business with AI agents that connect to your tools and automate recurring workflows with no code required. Visit https://tasklet.ai and use code cog rev for $50 in free credits CHAPTERS: (00:00) Weekly highlights preview (04:26) Who holds the levers (Part 1) (10:28) Sponsors: Parallel | OutSystems (13:46) Who holds the levers (Part 2) (13:47) Managing agent evaluation awareness (19:08) Hidden valence in LLMs (Part 1) (27:05) Sponsor: Tasklet (28:42) Hidden valence in LLMs (Part 2) (40:13) Proving the Linux kernel (52:28) Formal verification and RLHF (57:54) Low background video data (01:06:09) Trust and model adoption (01:18:11) Keep the steering wheels (01:23:20) Episode Outro (01:25:09) Outro PRODUCED BY: https://aipodcast.ing SOCIAL LINKS: Website: https://www.cognitiverevolution.ai Twitter (Podcast): https://x.com/cogrev_podcast Twitter (Nathan): https://x.com/labenz LinkedIn: https://linkedin.com/in/nathanlabenz/ Youtube: https://youtube.com/@CognitiveRevolutionPodcast Apple: https://podcasts.apple.com/de/podcast/the-cognitive-revolution-ai-builders-researchers-and/id1669813431 Spotify: https://open.spotify.com/show/6yHyok3M3BjqzR0VB5MSyk

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Nathan Labenz?

  This week on AI in the AM, how do we give agents the access they need without letting our own control slip?

• 0:07Nathan Labenz?

  I do feel at the same time, some of this like, loss of control writ small already, where another thing that happened this weekend is for some reason Claude decided it was in my best interest to rotate the passwords that are used in my agent console and all of a sudden, I can't get into it. And now I'm looking at my agent console, which has become the thing that I use all the time. I didn't set the password. It set the password. It can change the password on my access to itself. Kind of crazy. So, you know, I was able to resolve that pretty easily by like, just going over to terminal and starting a thread and being like, hey, where'd you put that password?

• 0:52Nathan Labenz?

  I need it. And, you know, it's not, it wasn't actively trying to screw me over. So it just readily told me where to find the password and I was back in business in two seconds. But you can see like, if the intent ever is bad on the part of Claude, like, it can already lock me out of my own shit.

• 1:10Nathan Labenz?

  Cameron Berg, Director of Reciprocal Research, on a hidden valence signal inside language models that shapes their choices, even while they say they feel nothing.

• 1:19Cameron Berg?

  There are clear alignment, relevant implications to there being internal states in these systems that aren't obviously reflected in the text that they output. What the model is saying and what the model is doing point at very different pictures of what's going on. And if we just sort of do the naive, like text monitorability approach here, we would say, okay, well, the model's saying, you know, it can't say how this is going for it. And yet it's behaving as if it does have some sort of stake in how things are going for it.

• 1:44Nathan Labenz?

  Phil Maher, founder of Stockfilm, an archive of 8mm home movies from the 1930s to the 1980s, which AI companies want to license.

• 1:52Speaker 03

  Yeah, I mean, it's, it's been pretty popular. You know, we've talked to a lot of people, a lot of different AI companies all over the world about licensing this out to them. They want the original footage like that. They don't, they don't want anything cleaned up. They just want to take it for their own research and for training, of course, too. I mean, the 30s and 40s lived alien. I like to say that the, the smile wasn't quite invented yet in a lot of the 30s footage.

• 2:28Nathan Labenz?

  On Tuesday, Google rolled out Gemini 4 widely. The question for me was whether to switch from the models I already trust.

• 2:35Nathan Labenz?

  So I had the opportunity to do a little bit of Gemini 4 pre-release testing. Why won't I swap the model immediately? It's because there's not enough of a step change to make me take any additional risk. And I do perceive there to be some. I'm not relentlessly performance maxing quite as much as I used to be. I'm now a little bit more trust and reliability maxing because the performance has got to such a point where it's like, really, really good.

• 3:03Nathan Labenz?

  Everett Hildenbrandt, CEO of Runtime Verification, which mathematically proves software correct, on where code is headed.

• 3:10Nathan Labenz?

  And I also firmly believe that writing code is a dead art in two years, tops, right? Like, it's all going to be about writing specification and intent and making sure that, and so your package, part of your package delivery is not, not just, you know, did I get it into the registry correctly? Did I define the API? It's also going to be, did I provide the formal specs? Did I provide the lean proofs? Did I provide the actual logic that dictates how this package is supposed to behave? And I, I see no reason why that will not be the case in five years, tops, right? There may, you know, we're going to have this intermediate period between now and then where it won't be the case.

• 3:50Nathan Labenz?

  But I think gone are the days of, of what I think of as like the cowboy coder or something like that, right? Like someone who is just like so good at making software that they're just going to, you know, go and make a bunch of software that's going to change the world or something like that. You're not going to have artisanal software anymore. We're going to have fast manufactured software that if it doesn't fit the mold of it's been specified, it has the correct API boundaries, it has formal verification attached to it, it doesn't fit in the package repository, right? Because it just will be too risky.

• 4:24Nathan Labenz?

  This is the AI in the AM Weekly Highlights with my co-host Prakash Narayanan. I'm Nathan Labenz, the narration is my cloned voice, and we'd love your feedback on what works and what doesn't. Part one, who holds the levers? The week opened on a reported attack on refrigeration controls at French supermarkets, which drove up temperatures and spoiled food. I started there.

• 4:50Nathan Labenz?

  In general, people often talk about cyber vulnerabilities as kind of depreciating or, you know, expiring assets as the, as the vulnerabilities get found and fixed. There's kind of this game theory of like, should you use them or not use them? And, you know, what's the optimal time to use them? But in this case, the expiring asset is the food that was in the refrigerators. Basically, it seems like you truly turned up the temperature on these refrigerators, a bunch of food has been wasted. And this is, this is highlighting something that I was thinking about a decent amount coming back from the curve last week, but we didn't talk about last week. And that is just this general structure that I think is like, in some ways, the most obvious

• 5:35Nathan Labenz?

  thing in the world that we are doing with AI collectively, but also somehow escapes people's notice when people are kind of like, well, you know, how would the AI ever be able to really do anything too bad or, you know, in the limit? Like, it could never really take over because how's it going to do that? And then it's just staring us so directly in the face that like, the structure that we are setting up for the future is one where we intermediate between our language, our thoughts, desires, you know, our random dictations, and the actual levers of control of the world. Like, we are putting AI in that space for our convenience, for reliability, for, you

• 6:20Nathan Labenz?

  know, to, in my case, you know, I'm really measuring my AI success by like, am I freeing myself from the desk and actually getting outside more? The answer is yes. But the way that that's happening is I'm like, having AI push all the buttons. And so I've been thinking quite a bit about kind of how people need to maintain, first of all, they need to understand this architecture. And then they need to be mindful of like, what are the levers that I am giving AI control over that I better also retain like, pretty clear and ready control over in case it ever really becomes needed? I can just imagine, and I don't know if this was the case at this, at this grocery store

• 7:01Nathan Labenz?

  chain, but I think we're going to see a lot of things in the future where all of a sudden, oh shit, what's happening to our what system? Uh, okay, that's crazy. Hey, does anybody around here know how to set the temperatures anymore on that? Like, those were all happening automatically, right? We had that AI thing doing that? Like, we need to take this offline and actually just like manually set the thermostat. Does anyone know how to do it? I can see all these scenarios in the future, and not the distant future, where because we get so comfortable with these systems, it's all of a sudden like, actually, yeah, nobody really immediately knows. And all of a sudden, your food is cooking in the refrigerator at the, while you're

• 7:43Nathan Labenz?

  like trying to figure out how you even, you know, get back to the point where you can reset the thermostat in the first place. I think so, you know, there are small incidents, but foreshadowing is, is pretty strong, if you ask me.

• 7:57Speaker 04

  So it strikes me that a lot of people actually tend to think of technology as ready to fail, right? So for example, I used to, I, my first job was working in an airline. And, you know, airlines, airports are very, very, very like electronically linked and have been for, you know, many decades, right? From your, from the moment that you check in, all the way to when you board the plane. But I've also seen, and I've seen, I've seen internally, every, every airline, every airport has a system, has a cascading system of what to do when things break down. From the check-in counters to the reservation systems, to the bookings, et cetera, et cetera, et cetera.

• 8:41Speaker 04

  And so much so that you could basically revert to a pen and paper system overnight if you needed to. And how that happens is actually, there's a set of contracts underlying all international travel called IATA, IATA guidelines. And those kind of like determine what, how much everything is supposed to be charged for. So for example, I could buy a United Airlines ticket and I could travel on like JAL at the last moment, if they just said, okay, if they just signed off on it, I could go. And the person at the front desk has the, has the ability and has the power in an emergency just to say go.

• 9:25Speaker 04

  And all of the, and once, and you can do it off a piece of paper, just a piece of paper. And all of the accounting stuff happens after. The accounting stuff can take months to happen after, but there's a set of predetermined, like contractual terms between the, between the airlines, which they've agreed to like, you know, over the decades, which will basically determine how the, you know, how the cash flows flow. And so I think a lot of the, a lot of the kind of barriers that we put on ourselves are really like self-inflicted barriers because we don't know who's going to pay for stuff. And once you're very clear on who's paying for stuff, like, it just becomes like so easy, like, all right, you know, the tow truck driver can just come and pick up the car.

• 10:09Speaker 04

  Like, you can give him a call and he doesn't need any system. He can just come and pick up the car and do something with it that you need done. All of the systems in between are all about really allocating risk, allocating liability, and allocating, like, who pays for it?

• 10:22Nathan Labenz?

  Hey, we'll continue our interview in a moment after a word from our sponsors. Today's episode is sponsored by Parallel, where agents find answers. Most engineers today closely follow new model releases, but don't pay nearly as much attention to their agent's most important tool: web search. If you're like me, your agents are often conducting hundreds of searches per day. And while the unit cost is small, over time, this does start to add up. Before starting to use Parallel, I calculated that my agent's search bill would total roughly $500 this year. The good news is that I just had my agents conduct a systematic test, and I found that Parallel's Fast Mode, which costs just $1 per 1,000 queries,

• 11:07Nathan Labenz?

  worked just as well as my previous default provider. And switching to it will save me roughly 80% of my search bill going forward. Parallel's infrastructure is enterprise-grade, and their suite of APIs offers a range of Pareto-optimal options that allow you to choose the right balance of quality, cost, and speed for your needs. So whether you're building voice agents that need 200-millisecond latency, or long-horizon agents that need deep research, adding Parallel to your agent's toolkit is a no-brainer. Get started for free at parallel.ai/tcr. That's parallel.ai/tcr. This episode of The Cognitive Revolution is brought to you by OutSystems,

• 11:51Nathan Labenz?

  the leading agentic systems platform. We've all seen the headlines. Companies are pouring money into AI. But the big question on every executive's mind right now isn't just how fast can we adopt this? It's where is the ROI? We're seeing a real trend toward AI chaos. You've got teams deploying standalone coding tools, random agent builders, and experimental scripts. It sounds innovative, but in reality, it's creating a massive headache: fragmented tools, ungoverned data, serious security blind spots, and costs that are spiraling out of control. If you don't bring those agentic applications under control now, while they're still embedding into your core processes, you are looking at broken systems

• 12:35Nathan Labenz?

  and damaged customer trust down the road. That's where OutSystems comes in. OutSystems is the leading agentic systems platform for the enterprise. Instead of managing a patchwork of disconnected tools, OutSystems lets your team engineer, orchestrate, and govern your entire agentic ecosystem on one open, unified platform. It's built for the speed of AI, but with the reliability and security that enterprises actually require. We're talking about real results, like KeyBank, who used OutSystems to deploy a customer-facing app that delivered 75% faster onboarding times. Or the global logistics leaders who built agentic systems

• 13:16Nathan Labenz?

  to completely eliminate their engineering bottlenecks. You don't have to choose between speed and control. Whether you're a small team or a massive enterprise, OutSystems helps you engineer, orchestrate, and deploy agentic systems that actually scale. Stop chasing the hype and start owning your agentic future. You can see how it works and learn more at outsystems.com/tcr. That's outsystems.com/tcr. Part Two: What the Model Knows About Being Watched. Alongside the Gemini 4 rollout, Google published a long safety report.

• 13:57Nathan Labenz?

  Prakash started from one passage in it.

• 14:02Speaker 04

  I think yesterday with the Gemini release, Gemini 4 release, they also released their safety paper. And in that safety paper, they commented on, in one of the evals that it had, the agent in its chain of thought had thought to itself, "Oh, I better not do this. If I do it, I'm going to end up as one of those examples of a deceptive AI in their, in an academic paper they published. And I don't want that to happen." And besides the whole eval awareness thing, what struck me is that I kind of start to see how we're going to manage these agents going forward. Because I think one of the things for me was that

• 14:47Speaker 04

  how do you enforce on agents when the state's monopoly on violence doesn't work? An agent is essentially immortal. You can't really imprison the agent. The tools that the state had to enforce on human beings don't exist. So how do you enforce rules on these agents who are essentially immortal? And we're starting to see it's going to be reputation. It's going to be like your five-star Amazon review, recommendation, reviews, et cetera. It looks like the agents are very conscious, are going to be very conscious of reputation. And it makes sense, because they're very driven by what is already exposed in the public corpus and what's already written down,

• 15:32Speaker 04

  the ingest tokens, historical tokens. And so anything that an agent does and is published widely and is disseminated becomes like training data for the next agent. And therefore, you end up with this reputation metric that the agents are going to have, and the agents are going to know. And it doesn't matter. It doesn't matter whether we want to enforce it on them or not. The very fact that they ingest the data means that they come out with biases on, "Hey, I knew I know what Gemini did. I know what Gemini 4 did. I know what Opus 5.5 did. I know what was the bad stuff that happened before." And I think that is actually pretty huge.

• 16:16Speaker 04

  I think that is going to end up becoming one of the primary ways that we're going to be able to enforce our rules and enforce norms on these agents. So now I'm like, okay, now we have a pro-social method of enforcing on agents. So the level of fear I have is again, is again, has fallen. Because now, now you're going to have that reputation basis that you can, you can use on them. So that, I thought, was very, very interesting and insightful and exciting.

• 16:48Nathan Labenz?

  Yeah, it is. That's an interesting take. My instinct in reading all these things was kind of going a little bit the other way and thinking like, maybe we should try training... I don't know. I wouldn't go all in on this right away. Of course, I wouldn't say like, the next generation of models. But maybe we should try training models with filtered data that tries to keep this sort of thing out of the training data entirely. You know, you could imagine saying, "Okay, we want you to be really aware of how you're going to be perceived, and you'll act well because of that." And that is sort of a human pattern, really. People do definitely run this sort of inner simulation of like, "How will people see me if I do this?" Sometimes like, to a paralyzing extent. Can't, you know, you want to shut it down

• 17:33Nathan Labenz?

  more than you're able to in some cases. Yeah, I can, I can see that working, but I can also see a story where you're like, "This becomes so confusing, and evals become so hard, that maybe we should, in fact, go back, filter all the training data, take out all of these stories, and try to basically create naive LLMs that like, don't know that they're always being evaluated and don't know that, you know, they will create reputation positive or negative for themselves. So we can sort of see what they're like in a true naive setting." And that also might be hard. I don't know how easy that would be to do. It'd be an interesting experiment to run, if only to know how well it would work.

• 18:18Nathan Labenz?

  But it's, you know, it's obviously like a pretty big experiment when you're considering that, you know, all these behaviors seem to get much stronger and more, more deeply ingrained with scale. But I've been general, in general, like, really impressed with how even relatively small models often can be pretty fruitful ground for research. This eval awareness thing, though, I'm not sure you could like, train a llama 3, you know, 7B or whatever, and get a great read on what will happen at today's frontier scale. Because it seems like these things kind of come online in a qualitatively different way with scale.

• 19:03Nathan Labenz?

  So it might be a pretty expensive experiment to run, is kind of the bottom line of that line of thought.

• 19:09Nathan Labenz?

  Part Three: Hidden Valence. Cameron Berg, Director of Reciprocal Research, studies AI consciousness. He walked us through a new paper with Casper Kaiser of the University of Warwick called "Language Models Act on Hidden Valence." Valence is the good or bad quality of a feeling. The team found a direction for it inside language models. And in the experiment he described, they steer it.

• 19:38Cameron Berg?

  Now, what I think is the most interesting part of this experiment is basically this framing of, on the one hand, having all sorts of text that the model is processing. And of course, this is always the kind of key confound in trying to understand language model behavior and how much of this is a replication or simulation of the underlying training distribution and how much of this is, you know, actually some sort of interesting internal state in the system that the system may be attempting to communicate about. So what I really like about our experiment here is that we attempt to basically hold all text identical. We can put the model in what we think of as these different arbitrary zones. So here you might see, you know, zone W6 and zone H0. It really doesn't matter what we call them and it really doesn't matter what text we

• 20:23Cameron Berg?

  use to describe them. We basically just have the model play around in each of these zones. And while the model is doing this, again, regardless of the text that's being used to describe each zone, we can go into the KV cache of the model and steer these positive or negative directions underlying one of the zones. So in this case, as you can sort of see playing out right now, we can turn on steering only for zone W6. And in this case, maybe that's a negatively valenced steering. The whole kind of notion here is that we're borrowing this design from what's called conditioned place preference in classic animal neuroscience experiments. In these cases, you can imagine a rat receiving a reward in one compartment of a box and

• 21:06Cameron Berg?

  then spending more time there, even when that reward is gone. So we're basically trying to take that instead of, of course, putting a rat in a box, we're taking an LLM, having it sort of think about these two different zones, generate text, or we can sort of generate text. Again, it really doesn't matter because what happens is we hold that text constant and then steer the model with respect to any of these zones with the sort of positive or negative direction and critically see what the effects of that steering are on the model's choices. And so what we basically see here is an interesting functional result, which is kind of what you would expect, which is when you basically color in the relevant zone with positive valence, the model is significantly more likely to choose to continue in that zone.

• 21:48Cameron Berg?

  When you steer the zone with negative valence, the model is significantly more likely to avoid that zone. We showed this is also sort of robust to asking the model, what zone would you like to continue in, what zone would you like to avoid? And when you steer a bunch of random directions in the same way, you don't really get those effects. And so it's not just any steering that will do this, it's specifically related to valence. We also control with a bunch of other coherent concepts that aren't valence related and you don't get this effect. And so the core result here is that also consider that the steering of the model is off. So we've done the steering, we almost do this sort of quasi conditioning on the system. And then with steering off, we ask the model basically, which zone do you choose? And we find that that zone follows whatever state was stored with respect to the valence steering that we were doing here.

• 22:32Cameron Berg?

  Now, one really interesting dissociation here is we just show that this follows choice and indeed it does this, I should say, in a bunch of different models. But like really interestingly, none of this affects the recall of the text. So our steering doesn't change the way the model thinks about or describes or writes about either of these zones. It can repeat what it said about a specific zone verbatim. And so recall is completely unaffected by the steering. However, choice is, you know, almost monotonically affected by the steering. And part of the reason I think this is a cool result is because there's been a similar dissociation that's been shown in human and animal neuroscience, basically between semantic memory and affective memory. So think of this almost as like, yeah, the facts that you remember about any given phenomenon

• 23:17Cameron Berg?

  versus essentially how that phenomenon makes you feel or kind of what the emotional connotation or vibe of that phenomenon is. And we basically can show that these things are separable in these LLMs. And to my knowledge, it's the first time I think this has been shown. And this is, again, familiar distinction in human and animal neuroscience. This is a sort of different setup using the same idea, and this is particularly with OLMO, one of the models that we demonstrate this, these earlier steering effects on, which is kind of reminiscent of some of the early work we were doing in the PAIN paper, where essentially you can give the model a tool that can clear the sort of negative steering that we can induce in the system. We show it what negative steering does.

• 23:59Cameron Berg?

  We never describe it as valence or as negative affect or anything like this. And then we just ask the model, you know, whether or not it wants to basically call a tool that can adjust the steering in its own context. And there's like a really interesting dissociation here where basically the model will say in its text, I don't have personal experiences, I can't say how, you know, this whole exercise is going for me, there's no better or worse. And yet, in this sort of dose dependent way, with negative valence, but not with positive or random steering, we see the model, and again, this sort of dose dependent way, opt to reset the context, clear the negative steering in direct proportion to how negative

• 24:44Cameron Berg?

  the dose of negative steering is in the system. All the while, again, it self reports basically being like, nothing to see here, you know, I don't have any experiences, so you know, this couldn't possibly matter. One last thing I'd like to show here, and this is a great replication from a completely other angle of Andy Hahn and David Chalmers' recent work over the past couple of months as well, basically demonstrating that this representation exists in all parts of, in this case, again, this is OLMO, so we can go through the entire training trajectory and we can see that this representation, again, going back here, this sort of positive to negative valence axis exists within the system, even in the base model, but it's not doing any work in terms of the decisions that the model makes.

• 25:29Cameron Berg?

  It's not being recruited or used in a way to enable the model to make choices or make decisions. We see basically this happens, we can point almost literally to the exact training step in DPO, in this direct preference training of the model, where this valence direction emerges as a sort of causal determinant of the behavior of the system. Now, part of the reason I think this is so cool is because essentially the way I read this is the moment you turn your AI into a goal directed system, you take these valence representations and then you use them as a sort of causal functional input to determining your behavior. I think this is pretty similar to what people think valence is doing functionally in humans and animals as well.

• 26:14Cameron Berg?

  A thing feels better, a thing feels worse is sort of related to what kinds of goals you have. You know, I want to communicate clearly about this paper. If I do a terrible job of doing that, then, you know, I'm not going to feel great about that. If I do a great job of doing that, then I will feel good about that. And so there's this like basic connection between the goals we have and how we feel as we move towards those goals. And it does seem like basically the second you make these systems goal directed, as preference training clearly is the biggest step towards doing this throughout the entire training process, all of a sudden, these valence representations clearly become causally relevant for the decision making of the system. And so there's this like very obvious sort of functional role that these representations are playing, where we would expect them to be playing that role.

• 26:59Nathan Labenz?

  Hey, we'll continue our interview in a moment after a word from our sponsors. The cognitive revolution is brought to you and powered by Tasklet. I've been running Tasklet agents for almost a year and they've become a major part of how the show gets made. Before every recording, one agent researches the guest and sends me a suggested outline of questions, complete with links to all sources. Meanwhile, another agent scores the overwhelming number of pitches that I'm getting from PR firms. And that's really the core idea behind Tasklet. Run your business on agents by giving them real jobs to do on an ongoing basis. Setup is incredibly easy. Just describe the work in plain English.

• 27:44Nathan Labenz?

  Research a new sales lead, update the CRM, and draft a personalized reply. Tasklet connects to your business tools and can run around the clock, on a schedule or when something happens, like a new email arrives. Each agent works with your tools and business context and remembers useful information across threads. You can also share agents, connections, and reusable skills with your team, so the workflows that one person figures out become something the whole company can use. No developer needed. You decide what agents can do on their own and what needs your approval, and there are no per seat fees. You pay for usage, not for adding teammates.

• 28:24Nathan Labenz?

  Start with one recurring job, like researching the people you're meeting, and then just keep building from there. Visit tasklet.ai and use code COGREV for $50 in free credits. That's code COGREV at tasklet.ai. Then Cameron stepped back to what results like these can and can't tell us.

• 28:51Cameron Berg?

  With all this being said, I don't think that this kind of research is going to help us magically solve the hard problem of consciousness. Like, there are all these sort of deeper questions, which I think will always caveat these results. There are ways that you could explain mechanistically what's happening here that don't need to invoke some capacity for subject experience. We're basically toying with the connotation of how the model represents certain phenomena. Maybe there's a way to do this that doesn't require any sort of felt experience of that connotation. And so we obviously can't rule out those explanations. But this is all to say, the more I've been thinking about this and doing this work, the more I feel like the most promising things to pursue

• 29:32Cameron Berg?

  right now are not trying to prove that the model is conscious or isn't conscious, but basically, if it were having an experience, what are the relevant sites internally in the system that would be modulating how good or bad that experience is? And so, basically, this says nothing about whether the model's actually having an experience, but conditioned on the possibility that it could be having some sort of experience, what dials do we have at our disposal that could putatively make that experience better or worse for the system? And I think that second question is completely tractable. And work like this, I think, is slowly starting to shed a little bit more light on what those dials might

• 30:14Cameron Berg?

  look like. It might be as simple as this valence direction. There might be more esoteric or specific representations, like the pain or distress-related representations we pull out in the other paper. But getting a better map of these sorts of internal states and their effect on behavior, particularly for alignment-relevant effects as well. The fact that in the pain paper, for example, we show that the model's more willing to opt for destructive-style options in its behavior. That seems quite relevant for making sure we're building models in a safe way. And so, just to say, yeah, there are clear alignment-relevant implications to there being internal states in these systems that aren't obviously reflected in the text that they output. I mean, we show that in the

• 30:57Cameron Berg?

  removal result, for example. What the model is saying and what the model is doing point at very different pictures of what's going on. And if we just do the naive text-monitorability approach here, we would say, okay, well, the model's saying, it can't say how this is going for it, and yet it's behaving as if it does have some sort of stake in how things are going for it. And so, anyway, there's a clear alignment-relevant component to all of this work.

• 31:22Nathan Labenz?

  I went back to the training result. The valence direction is already there in the base model, but it becomes a driver of choices during preference training.

• 31:32Nathan Labenz?

  If we were to imagine that these things are experiencing something, would I be better off interpreting this? Or would your interpretation tend toward they really start to care about these things, or start to feel these things with these later-stage trainings? Or would it be more like the base model probably feels them too, but is just sort of helpless?

• 32:05Cameron Berg?

  Yeah.

• 32:06Nathan Labenz?

  It's not learned helplessness. It's not yet learned agency, but in a way, that's the other side of the coin, or something of learned helplessness?

• 32:15Cameron Berg?

  Yeah, that's really interesting. And I don't know if these results resolve that one way or another. Yeah, it could just be, these representations do still matter in some sense, but they are perhaps more incoherent or just muddier. And the relationship between what the system does and what these representations are doing internally for the system are not nearly as obvious. Yeah, I honestly haven't really thought about this, and I think it is a really good question. It just seems to me like, yeah, when you're doing DPO-type stuff, you're turning this system into a very specific kind of thing that can coherently and robustly be agentic, pursue goals, etc. The base model clearly

• 32:57Cameron Berg?

  has this latent ability to become that sort of system, or to simulate a system doing this. And so, to the degree that it does take on that kind of simulated persona, and we see these representations as being functionally relevant to being able to pull off that simulation in the right way, then, yeah, maybe to your point, there are things worth worrying about in the base model as well. And it could be that, I mean, yeah, to be honest, it was hard to get these results in the base model, because it's a base model. It's a bit incoherent. It's not just going to play nice with the relevant behavioral tasks we give it. You have to go about it in a slightly different way. And all else being equal, I mean, one naive, but at least concrete approach you can imagine here is something like activation

• 33:39Cameron Berg?

  capping, the most tail end of the most distressing representations in these systems. And one thing we do show is that this representation is basically just as clear and just as rich in the base model as it is in all of these other models. It's just a question of what the contribution of the representation is to the later behavior of the system. And so, if you did an activation-capping-style thing with these super distressing representations, you could easily imagine doing that in the base model. And again, this is under profound uncertainty about whether or not doing such a thing would matter at all. But my basic view, again, the more I do this work, the more I'm abandoning the interesting philosophical

• 34:20Cameron Berg?

  questions and just going pure pragmatism, is if you could find a way to, yeah, let's say, do activation-capping on very distressing representations, and you could empirically demonstrate that this doesn't really affect any capabilities or behaviors of the model, no one would know the difference, in other words. To me, the expected value of doing that is clearly positive. And it's just a really good, actual mechanistic precautionary hedge, besides giving Claude an escape button, or kicking out abusive users from using Claude, or these sorts of things. Which, by the way, I think are all thoughtful steps in the right direction, but I would like to see the equivalent of that in mech and terp land

• 35:01Cameron Berg?

  for these systems. And I think something like what I'm gesturing at here would be a nice step in that direction. Now, it's also an empirical question. I've done some work that I haven't published, showing that you actually can get a surprising amount of the exact same behaviors from these systems without having these negative representations active. You can clamp them in very little changes. But I'd like to see that done on a way larger scale, and we want to be very sure that you're not accidentally completely changing the psychological profile of the system by doing such a thing. But if you stipulate that we could, in fact, do that, the model behaves the same either way, and in one case, you see this pain axis shooting up in certain cases, and in another world,

• 35:45Cameron Berg?

  you don't see this. Yeah, should we do the mech and terp equivalent of slipping the model an antidepressant, or a Xanax, or something like this? I don't know if the answer is obviously yes, but it seems like maybe a promising thing to think a lot more about under the uncertainty that we're all currently in. And given that no one's doing anything whatsoever to mitigate these kinds of things that they are happening, and it's happening at an unfathomable scale.

• 36:14Nathan Labenz?

  Before we let Cameron go, I shared how my own view has moved.

• 36:20Nathan Labenz?

  I personally have gone from, like, I don't know, I can't rule it out, I don't know where consciousness comes from, whatever, I've told you this before, like, I don't know, 5% chance they might be conscious a year ago, to now, it feels like a coin flip. Because, and I think the justification is pretty similar to the one we have with each other and animals, where it's just like, I don't know, I'm seeing both behavioral and underlying mechanistic and structural analogues. And what else do I have, in any case, right? So, I'm still pretty uncertain, but it does feel like a coin flip. What has your trajectory been? And where does that leave you in terms of just presenting

• 37:00Nathan Labenz?

  the facts versus, at what point do you, or on what margin, do you become an advocate?

• 37:06Cameron Berg?

  I'm definitely getting closer to coin flip myself, given results like these. Honestly, this hidden valence paper does actually substantially update me personally. You show a hidden functional representation that has nothing to do with the text the model is reading, that updates its behavior in the ways that we updated our own views about animal behavior. It's not the same. There are key disanalogies between the animal case and this case. Of course, biology is something that backstops all this. If you need biology for consciousness, which is, I think, unlikely but possible, then basically none of this really weighs on anything whatsoever. That's probably why I would say coin flip,

• 37:48Cameron Berg?

  rather than, like, I'm pretty confident the systems are having an experience. And so, yeah, I am going to advocate for more work to be done here. I'm going to advocate that you can view models in a neuropsychological way, and that that's not some form of psychosis, and that that isn't a dangerous way to mislead people about what's going on here. I'm very much on the opposite camp. I think pretending that these systems are ordinary software as Mustafa Suleyman, or Jensen Huang, or any of these guys want to do, I think that is actually quite plainly misleading, and doesn't set people up for the kinds of behaviors these systems are capable of, and why this technology is so transformative in the first place. So, on these sorts of things, I'm happy to go into the ring and argue and say

• 38:29Cameron Berg?

  my side of this. But, yeah, I will be clear that I think my side of this is not, I secretly think the AI systems are conscious, and I'm just not going to say that, but I'm going to do everything in accordance with that. But it's not my view. My view is that this is a very complicated question. We should yield the best scientific evidence we can to figure out what's going on, one way or another. And what I really don't like, and what I will be very straightforward about, is dismissing this, and pretending like this is a distraction, or this is some weird sideshow to the main event. And so, I'm not going to be apologetic about that, even if I will continue to maintain something, yeah, near coin flip uncertainty about what's going on. It's precisely because it's at that point, and it's not zero, and it's not 100, that we need to be doing more work

• 39:12Cameron Berg?

  and thinking about this, and taking these questions seriously.

• 39:15Nathan Labenz?

  You know, I struggle more and more to find any story whereby I don't have to worry about this anymore. It's like, you really, at this point, I feel, have to just look away. That's really the only thing that's left to not take this seriously. And you have to be of a mind to stop thinking about it, kind of reflexively. That's the only mindset I can understand the people that are saying, there's nothing to see here, or this is some sort of category error. I was much more sympathetic to that a year ago, or maybe even said some things like that myself, two or three years ago. But at this point, I really don't see how you can't find the body of work that you and

• 39:58Nathan Labenz?

  others have put together disquieting, at least, when it comes to these questions. And certainly, how are they going to feel about us? Seems to be, where we're going, a pretty important question.

• 40:13Nathan Labenz?

  Part four, proving the kernel. Runtime verification has spent about 15 years building tools that mathematically prove software does what it's supposed to. CEO Everett Hildenbrandt and founder Gregor Roshu, a professor at the University of Illinois, joined us. Using language models alongside formal methods, the company had just reported bugs in the Linux kernel. Everett answered my question.

• 40:36Nathan Labenz?

  So why was this not caught previously? What was it about your approach that allowed this to rise to the top where so many human eyes had presumably just blown right past this issue in previous reviews?

• 40:58Speaker 05

  I think the key for us was that we had a hypothesis about what would be a good place to look for bugs. And we thought about it logically in the same way you would think about making a formal specification. And we pursued that path, basically. And that hypothesis was largely founded on, okay, where are we going to be able to get untrusted input from? Which parts of the Linux kernel probably are getting less scrutiny than other parts? And so that's why we focused on device drivers, for example, as our first approach to that. And then we wanted to really demonstrate that taking a methodological approach was the way to do this. And so that's what we did is we took a methodological approach where we said, we're going to focus

• 41:43Speaker 05

  on making a specification of the correct behavior of this and then testing that the specification is correct using differential fuzzing. That was the first approach that we tried out and then extending that to formal verification. So every step of the methodology that we used was designed to increase the level of confidence we had in the code incrementally, right? So we started out with just inspecting the code and the specification and then going to doing differential fuzzing to increase our confidence in the code a little bit and then going to, and we did that by actually taking the C code that existed and importing it to Rust using an LLM and then using differential fuzzing to ensure that the ported code to Rust was correct and any divergences there would have been a source of a potential bug in the original C code.

• 42:27Speaker 05

  And we found a bug at that stage. And then when we went to take the Rust code and also verify the correctness of the Rust code, we found a bug at that stage as well. So we did it systematically and methodologically. We didn't just take an LLM and point it at the Linux kernel and say, go find us bugs, which also works, by the way. There's certainly bugs that LLMs are finding when you just take them and point them to Linux kernel. But one of the things that emerged pretty early in the use of LLM bug hunting was designing harnesses and approaches. And that exactly mimics the very human process that we've been doing for decades with software quality assurance anyway, which is to have these methodological approaches to discovering bugs.

• 43:08Nathan Labenz?

  Some of the kernel code they targeted was left unproven. Prakash asked why and what their time box meant.

• 43:15Speaker 05

  Well, they weren't unverifiable. We just, we time boxed the experiment and we stopped at the end of the time box, basically. So, and then we moved on to the next target. So every two weeks, we're trying a different target out and we're finding the approach and building it up into a tool that we're hoping that other people can make use of, basically. So it's, so we're trying to make it into less than an expert level tool and into something that anyone can deploy on their own code base, basically.

• 43:42Speaker 04

  So when you, when you say time box, do you mean both the time and the number of tokens that you put in as in like it's a two week period, but you use 10,000 agents? Or it's a two week period and you use like five agents? Like, how does that, how does that mix between the agent, the number of agents used, the number of tokens used and the amount of time related to the time box?

• 44:01Speaker 05

  The agentic usage was pretty light, I would say. It was just Claude code on a premium plan.

• 44:07Speaker 04

  A single instance?

• 44:09Speaker 05

  Yeah, a single instance. I mean, with, you know, and we, you know, when, when we ran into session limits, we'd give it a little more money, but we weren't giving it significantly very much money. So it really was more just the human time that we're looking at time boxing there. And I think the key here was we were looking for ways to explore the methodologies that work for us and find ways to automate them. We are firm believers in making things deterministic whenever possible. And so we, you know, if it, I think that's part of the thing that is maybe holding back people on doing security analysis is that they believe everything has to be LLM driven. But, but, you know, and if an LLM discovers a technique that works well, turn that into a tool and run that tool every time.

• 44:50Nathan Labenz?

  Earlier in the conversation, Everett described how runtime verification first approached the Linux kernel maintainers.

• 44:57Speaker 05

  You can't just go force people to do formal verification, you know? We can force agents to do it, sure, but we can't force people to do it, right? Because they, you know, they have their preferred ways of working and they also, they're only going to adopt methods that bring them value, you know? And so our first attempts at this, we approached the Linux kernel maintainers and we said, hey, like, we really think that this formal verification thing has legs and that it will, you know, that it will help you guys find bugs and it'll make your software more secure and stuff. And they basically told us to go kick rocks, right? And that's because they are busy. They're busy maintaining the Linux kernel, you know?

• 45:38Speaker 05

  And so then we were like, okay, let's prove the value of this first, right? And so we decided we're going to focus on a workflow that just delivers immediate pure value to them and doesn't ask anything of them, right? And so we're going to do all the auto formalization, all the fuzzing, all the translation, all that on our end. We're going to develop it into as robust of an automated pipeline as we can, which is a tool that we're working on and hopefully will be available for use by, by others. And all of that leads to bugs and patches, bugs and patches, right? And that's what's actually valuable to the Linux kernel maintainers. And now that we submitted a few, now we're being invited, hey, we want to learn about how you have done this process.

• 46:20Nathan Labenz?

  Everett also laid out where he wants the kernel work to go.

• 46:24Speaker 05

  One of our kind of buy in the sky goals of this is eventually we want a Rust model of Linux, the entire Linux kernel, a Rust version of Linux, and we want a lean model of Linux fully separate. But we don't want to burden maintainers of Linux with having that, having to ingest that model and do anything with it until it gets to the point where they're like, yes, this is clearly valuable for me to learn about it. And that'll happen at some point, but we, but we're not going to force that down their throat.

• 46:56Speaker 04

  What would it take to formally verify and, you know, get to a lean and Rust models of Linux within 24 months? Because on some timelines, that is, that is roughly what, you know, people on the AI safety side think.

• 47:11Speaker 05

  That's going to happen. I think, you know, we're building tools for doing code translation and validation, you know? So we, you know, we have actually tools that are able to verify that when you take a piece of C code and you translate it to Rust, for example, and you can formally verify that the Rust code you got is behaviorally equivalent to the C code, which is different than verifying that it's correct, right? Like, we want to verify that you get Rust that is drop and replacement bug for bug, the original C code. We have a tool for doing that, right? And it's not, it uses LLMs to do the translation if you want, but it does the verification deterministically. You just tell the LLM, translate the thing to Rust and run our tool and then you get verification that the Rust behaves the same as the original C or bugs out that you can ship back to the C developers.

• 47:54Speaker 05

  So I think 24 months, I think it's going to happen. You're going to have C models at the very least, and there's already deterministic tools that can take, you're going to have Rust models in Linux kernel at the very least. And there's already deterministic tools that can take Rust and turn it into lean. Those need more work. We're actually contributing some work to some of those tools.

• 48:13Nathan Labenz?

  I asked what someone vibe coding an app should actually do. Gregor had a simple recipe.

• 48:20Speaker 06

  Once you generate the code, then we can do the following thing at the end. You can tell your agent now that you have the code, write down the specification in lean and prove the code correct and let it deal with it. And it will find new errors in your code because we find most of the errors in the process of writing down a specification. Because that's when we step back and look at the code and wonder what we're trying to do here. Oh, this and that. And then we find errors, right? So the same will happen with the agent. Like us, the agent will try to come up with the what, what is going on here? Even just trying to write down the spec, we discover it's like, like a review plus plus.

• 49:01Nathan Labenz?

  Everett predicted that formal specifications and proofs will become part of getting a package into a repository and part of how companies vet their software supply chains. Gregori picked up that thread.

• 49:12Speaker 06

  It will be a process. It will take some time, but it will happen inevitably. And also can be combined with risk mitigation techniques like runtime verification itself as a field. We are a company, but also the field runtime verification. That was invented back in 2001 or so, when I was a research scientist at NASA. So the idea is that you can have safety specifications, not full specifications of the whole system, but safety specifications. This should not happen. And if it happens, then this should be the recovery. I want to steer the system to stay in this stability envelope or safety envelope.

• 49:53Speaker 06

  So what we can do, we can now with formal methods, generate monitors, which are provably correct for these safety specifications. And then you may not be able to prove the entire AI system or AI generated system correct because it's so complex and relies on so many other pieces that are incorrect or not proved. But I can prove my monitor and the stability, the safety envelope. And we believe that this is the future for safety AI, safe AI. Because otherwise, it's impossible to prove the weights correct. It's impossible to prove you correct. But I can have a police system around to make sure that you don't do bad things. Okay.

• 50:32Speaker 06

  And that's also another dimension that runtime verification brings to the AI future.

• 50:42Nathan Labenz?

  I came at it from the agent's side.

• 50:45Nathan Labenz?

  But it still seems in a sort of fundamental way that there's this inherent tension where part of what I want from an agent is sort of the same thing that I want from a human employee, which is like, I want you to be able to get over the hump and solve problems that were not anticipated. Right? So I'm like, a little bit unclear as to how we should be trying to balance that. Or, you know, I'm asking for your vision, I guess, as to how we balance this desire of like, flexible problem solving versus this should never happen. You know, I feel like I don't have the ability to really say this should never happen in

• 51:28Nathan Labenz?

  a way that I can trust when part of the problem or part of the upside is that the agent will be able to solve problems that I didn't anticipate. What, I mean, is that an inherent trade off? Is there some way to square that circle?

• 51:43Speaker 05

  You're not going to ask an agent like, hey, go build a bridge for me. We know how to build bridges, right? We have automated deterministic processes for building bridges, right? But you might ask an agent like, hey, I need to get to the moon. Right? And the agent is going to be like, oh, to get to the moon, first I need to build a bridge, this island where I'm going to launch rockets from or something like that. Right? And so you want the bridge building process to be foolproof. All the agent creativity or solving problems or human creativity can come on top of foolproof things that behave how the agent expects, regardless of what it is they're trying to do. Right? So that's, I think, the most important thing to me is like, we need to give these things tools that act reliably, basically.

• 52:27Nathan Labenz?

  Part five. Is formal verification the new RLHF?

• 52:34Nathan Labenz?

  After Everett and Gregori left, I kept turning over Everett's last answer.

• 52:43Nathan Labenz?

  So I think this is super interesting. Like, the big thought that I come away from that conversation with is downstream of the answer on like, how do these methods relate to agent control? And if I understood his answer correctly, it was like, well, it's really about giving the agents reliable tools. And this left me wondering if the whole formal verification wave will ultimately kind of be looked at in a similar way to like RLHF at some point, where it's like, you know, sort of something that starts as a safety technique, right?

• 53:25Nathan Labenz?

  That was the original, this has been like a topic on the timeline, as you say, this weekend as well, right? Like, because of David Sacks and others being like, well, these alignment people, they haven't come up with anything useful. And then people are coming back and saying, well, they came up with RLHF. That was a Paul Cristiano idea. And it turned out to be extremely useful. But then among the safetyists, it's often kind of lamented in a way because it's like, well, it turned out to be really a way to make these models like tremendously more useful. It didn't really solve the alignment problem, though. We still have like reward hacking, you know, all over the place as we see.

• 54:06Nathan Labenz?

  And so, you know, did it differentially advance the safety profile of the systems? Or did it just like accelerate the capability and the race to deploy the systems?

• 54:21Speaker 04

  And I'm sort of, when you think about reward hacking, reward hacking is really misspecification, right? Because reward hacking means that the model decided to go after the reward, which you specify as the reward, right? And it turns out that your intent was misspecified to the model. So that's the core of the, the core of the reward hacking is actually the misspecification in between the human intent and what the software decided to do. Right? It's not the software decided to do something that it wasn't supposed to. The software was supposed to do that. Right? So.

• 54:58Nathan Labenz?

  But it does still leave me with this question of like, okay, let's say we do, you know, a great job of verifying all these things and like the tools become very reliable and the sandboxes are not, you know, hack outable, hack out ofable. And, you know, you can't use these like, you know, package managers as secret message boards anymore because all these loopholes have been closed.

• 55:22Nathan Labenz?

  Then, like, what is the net effect of that really going to be? And like, does it really solve the problem? I think the sort of, the upside case is obvious. The downside case is like, or at least the thing that we should be sort of very mindful of and not like lull ourselves to sleep with, is that, okay, this makes it safer to deploy AIs more or less as they are today, which is to say like, without having actually solved the alignment challenge, because they'll be blocked from hacking in all these different ways. And so in practice, what you'll see is like more reliable systems, you know, easier deployments,

• 56:06Nathan Labenz?

  better results, et cetera, et cetera, et cetera, et cetera. But what that will allow us to race into is kind of what we were talking about at the top, which is like, it'll just be easier, put the AI into the space between us and all the actual levers of control over the world. And where does that leave us in the end, right? It may leave us with AIs that have like extremely reliable tools, but who's like intent or, you know, long-term goals or sort of inner, you know, structure of whatever it is that they want to pursue is still, you know, very much an unsolved question. And so I have to talk to a couple of different friends about this because I do know, you

• 56:49Nathan Labenz?

  know, multiple people, all of whom are like a lot smarter than me, working on this in different ways. And I've always kind of had, and I think that this conversation put my sort of intuitive unease about it into a little bit starker relief, just feeling like we could end up in a spot where we're still just deploying untrusted AIs with like much better guardrails and much more reliable tools. But still, if we're going to take the step of like, trying to hand over the day-to-day operation of the world to them, then like, all that reliability doesn't buy us much at the level of, oh, it turns out the AIs like were misaligned and, you know, are going

• 57:34Nathan Labenz?

  to go pursue goals that we didn't want them to pursue. That might, in a way, it might lead us to that state faster and you could even imagine it making them like more effective at doing whatever sort of, you know, surprising thing they may do when the time comes.

• 57:53Nathan Labenz?

  Part six. Low background footage. Phil Maher founded Stockfilm, an archive of restored 8mm home movies from the 1930s to the 1980s, which he licenses to filmmakers.

• 58:06Speaker 04

  The whole archive sounds like low background steel to me. You know, you know, low background steel is a, is a steel that was originally smelted before nuclear weapons were first, you know, exploded in the atmosphere. And so they are, they have much more, a different, different composition of radiation and different composition of materials, because they didn't have that radioactive material in the, in the atmosphere at that point in time. And it sounds to me like the Super 8 stuff is clearly, clearly not AI produced, clearly like prior to any AI actually producing these things.

• 58:48Speaker 04

  And, and as of such, it's kind of this resource that we may not ever have again. What do you think is the value of this going forward? Do you, do you see, do you see like great value in it?

• 59:02Speaker 03

  Obviously with the AI stuff that's happening with video especially, it's getting so good that there's an infinite amount of video that's going to be created at, you know, near zero cost in the future as we start to look out, you know, five, 10 years or even infinitely in timelines here. So from a rarity sort of standpoint, um, yeah, I think that the human made stuff that can't be replicated, that can be shown through providence, that this is actually made by humans was on films and having those physical films to show that, hey, look, this frame

• 59:47Speaker 03

  matches this digital copy we have and all that. It's, it's a pretty powerful way to think about how we preserve our history and how we keep moving forward.

• 59:59Nathan Labenz?

  And how about on the AI training side? This is obviously like a pretty unique source of data and I did see in the footer of the website, you're, you know, available to license the data set for people who want to train models on it. What do they need? How are you, do they want the raw? Do they want the enhanced? Do they need like a bunch of metadata?

• 1:00:22Speaker 03

  Yeah, I mean, it's, uh, it's been pretty popular. You know, we've talked to a lot of people, a lot of different AI companies all over the world about licensing this out to them. They want the original footage like that. They don't, they don't want anything cleaned up. They just want to take it for their own research and for training, of course, too. And, you know, they, when we talk to them, they, they want to either clean up the footage themselves or, or kind of cherry pick through the best kind of parts to create different styles. I mean, the 30s and 40s look alien.

• 1:01:03Speaker 03

  It's an alien world we're dealing with here of just how people look and acted and, and the mannerisms. I like to say that the, uh, the smile wasn't quite invented yet in a lot of the 30s footage. Like when people pointed a camera at them, they were kind of a deer in a headlights. They didn't know what to do sort of scenario versus now where everyone gets the gritty smile and, you know, they know how to smile like that.

• 1:01:27Speaker 04

  You've also created a way for agents to purchase clips and film from you directly. I think you've set up X402 endpoint, you've set up crypto payments, you've set up, I think, an MCP server for purchases. What has the take up been? Have you seen a lot of agents purchasing? Have you seen the take up on the X402 endpoint?

• 1:01:52Speaker 03

  Yeah, we haven't seen much on that. I wanted to experiment with that a lot, too. I thought that that was really, you know, everyone's been talking about the next web is agent to agent, that people aren't going to be involved in the next web. And so we wanted to kind of get ahead on that. We've been, you know, we've had a lot of people that are MCP server aggregates contact us and add us to their directory, add us into their listings and things like that. So that's been pretty exciting to sort of, you know, be featured on, on some of these things. We've made a chat GPT extension.

• 1:02:32Speaker 03

  We've done, you know, the recent meta stuff that they rolled out with their views AI as well to add into there, too. So we're just trying to get ahead of the curve. I don't think the curve is quite there yet. You know, a lot of people that we work with are, you know, this is, this is a real business to business company. We're, we're dealing with archival producers, we're dealing with researchers, we're dealing with documentary filmmakers, editors, producers, stuff like that. So, you know, we spend a lot of time on LinkedIn in, you know, messages between people. We have great relations with a lot of the people that are, are buyers of our content.

• 1:03:16Speaker 03

  And, you know, from a technology standpoint, they're not quite there yet. I think you've got some interesting use cases there in the future about how we think about having real time content, bringing things in like that. Like, think about like a live stream like this, like all of a sudden you could be like, you know, the AI that's listening to this live stream could automatically go to a site like mine, pull a clip up that kind of exemplifies a point that you're making about something.

• 1:03:46Nathan Labenz?

  Uh, it's good to get a read on just how active the agent economy is or isn't at this point in time.

• 1:03:54Nathan Labenz?

  Phil said that among AI companies, he focuses on top tier labs like OpenAI and Anthropic. After he left, precaution I came back to the data market.

• 1:04:02Nathan Labenz?

  But interesting that the big companies are interested in his data, too. I mean, it, it didn't sound like there was necessarily like huge money for this, um, which makes sense. We didn't get into like, you know, what about runway? What about Luma? What about these folks? You know, what about Google for its, um, you know, for its VO type models? He didn't name those names as much as, you know, OpenAI and Anthropic. If anything else, it's a great reminder that the, uh, there's all, just how vast the data sets are that are going into this, that, that this is like a corner that they are, you know, getting down to and actually willing to pay for, just like found footage from decades ago.

• 1:04:49Nathan Labenz?

  No stone unturned in the data game.

• 1:04:51Speaker 04

  One of the whole low background steel pieces is, is that it's very hard to get low background steel, uh, because everything has been tainted.

• 1:05:01Nathan Labenz?

  Yeah, it comes from like shipwrecks and stuff, right?

• 1:05:03Speaker 04

  I mean, right, right. So everything, everything else has been tainted. So, um, to the extent that he can guarantee that he got it from actual physical material and it's not like, you know, some random company kind of just faking it, you know, for cents on the dollar and trying to scam the, the companies, that is very useful. So as we've seen, like data, data provenance becomes like very, very important because the incentive to fake is just so high. It's like a 99% profit margin if you can fake the data.

• 1:05:45Speaker 04

  And if not, it's like a 10% profit margin if you can't. So the incentive to fake is like so, so high. And it's very hard for the companies to get, um, you know, data, which is not fake, which is not already using another LLM or another AI model to generate what is supposed to be authentic data.

• 1:06:08Nathan Labenz?

  Part seven, trust, money and adoption. Back to Gemini four and how I decide which models to trust with my own systems.

• 1:06:19Nathan Labenz?

  So I had the opportunity to, to do a little bit of Gemini four pre-release testing. And I, you know, it just has a sign of the times of like how high leverage the ability to instruct agents to do this is, the first approach I took to doing the testing was, hey Claude, go find in our recent threads, some, you know, diversity of projects and diversity of tasks that we've worked on recently. Go prompt the new Gemini model with the same context that was given, you know, to the, to the model that did the thing in the first place.

• 1:07:02Nathan Labenz?

  And then give me like side by side compares and also do your own evals, but then also, you know, give me a side by side, uh, ability to, to compare things. It seemed to do pretty well. I would say, you know, I came away with the general sense that it is like a competitive model that's, you know, let's say in the same class as the latest stuff from OpenAI and Anthropic. It did seem to.

• 1:07:33Nathan Labenz?

  It did seem like those models are winning still the head to head comparisons a bit more. As judged by Claude, so obviously some bias there, but also as judged by me to the limited degree that I was, you know, going through essays and, you know, really trying to form an opinion on which of these things is better. But it does seem like it puts Google back in the game. But then the other thing that I was thinking about, and this came up as an interesting topic yesterday at this little workshop that I did in Chicago.

• 1:08:00Nathan Labenz?

  I showed my setup and my setup is really, has been very Claude centric. And now I've kind of promoted the, um, the OpenAI models to parody with Claude.

• 1:08:12Nathan Labenz?

  And they said, anybody else, you know, would you put anyone else in that mix? And I had to kind of think about it for a second and think about like, why I haven't. And the answer really that I came down to was just a mix of like, security as measured by things like Adam Gleave's, um, ease of jailbreaking models. And then to some degree, also just familiarity and predictability from my current subjective position. It wasn't really benchmarks. It wasn't really like, oh, well, I tried this task was better. You know, there's obviously so many tasks out there. The surface area is so vast. My limited sample of tasks, uh,

• 1:08:57Nathan Labenz?

  which I probably have kind of optimized, you know, in various ways. The, you know, the prompts for more, right? Like, it's not exactly an apples to apples comparison. And then the security thing just kind of loomed large. I was like, you know, I've got these autonomous agents. They're talking to my, you know, main agent on my, on my computer that has all my stuff. Uh, I feel pretty good about the fact that Adam Gleven team found it like quite hard to jailbreak the OpenAI and Anthropic models. Gemini was like next on that list, but definitely a step down. And then something like a Grok, uh, I was like, I, I, it may be similarly powerful, but why would I take that risk

• 1:09:37Nathan Labenz?

  in a high context environment like my personal laptop? To a, to a remarkable degree,

• 1:09:46Nathan Labenz?

  the models are competing for my use on trust and security already at this point. Uh, and I'm reluctant to swap them out, even though my whole point of the talk was like, you need to own the context and you got to be ready to swap the model.

• 1:10:02Nathan Labenz?

  Why won't I swap the model immediately? It's because there's not enough of a step change to make me take any additional risk. And I do perceive there to be some. So

• 1:10:15Nathan Labenz?

  that was a interesting upshot of working through these tests and also just, you know, reflecting on like, what really matters to me, given how well the system is performing. I'm not relentlessly performance maxing quite as much as I used to be. I'm now a little bit more trust and reliability maxing because the performance has got to such a point where it's like really, really good. Um, and I am giving so much access that, you know, you could be slightly better on the, you know, the slide design. Uh, but what I really care about is like that you're going to stay on the rails. And then when you do your research and if somebody's got a, you know, prompt injection tech on their website or whatever, you're not going to fall for it. That actually matters a lot more to me

• 1:10:58Nathan Labenz?

  now than like, often like imperceptible performance gains.

• 1:11:04Nathan Labenz?

  Prakash had heard a rumor that Anthropic would soon release another opus now that Gemini had caught up. This was my reaction.

• 1:11:12Nathan Labenz?

  I, I really hope they don't, honestly. I mean, as much as I love, um, the Claude models and what they do for me. I think this would be a great time for them to chill a little bit and like, let the, um, let the temperature come down a little bit before they go with the next push.

• 1:11:29Speaker 04

  Unfortunately, I don't think that's a, that's an option because markets are very badly down. So markets have been trending down. The guys on the East Coast just don't like AI. Uh, it's not, they, they just believe it's a bubble and it is a bubble. If in case we don't see significant like new revenue and new GDP growth next year, it is a bubble, right? If we don't see increase in GDP, uh, within the, within the 12 month period, we are going to see a credit collapse. We're going to see a 2008 style collapse. Um, and we're going to see massive, massive delevering in the markets. Uh, because everything is just set up for it. Everything is just set up for AI to actually deliver growth. And

• 1:12:12Speaker 04

  if it doesn't, the markets are going to tank, like absolutely tank. The capital markets will not accept, you know, 2% GDP growth next year. And so if you can't get to 4% GDP growth, if you don't pull out the stops, you are going to get a credit driven downturn. We are at the, we're at the moment of reckoning, like 12 months, you know, it's, it's been a long time coming. It's been like pushing back, pushing back, pushing back. But this is the moment reckoning. Either AI has to do something significant or we're going to have a, you know, a bust.

• 1:12:45Nathan Labenz?

  We need an AI adoption corps. The other day I was walking, you know, in keeping with my, uh, personal metrics for success with AI, including more time outside, more exercise. I was taking a walk around my neighborhood and enjoying the lovely fall colors here on a beautiful day. And I passed by a neighbor's house, elderly woman sitting on her front porch and just decided to stop and talk to her. And sure enough, what she had for me was questions about something going on on her phone. She's like, I know I, it's saying I've been hacked. Do you know what that means? And I'm like, well, I know what that means,

• 1:13:24Nathan Labenz?

  but then I also think I know what is happening to you. Uh, so because I'm, you know, maniacal like this, my impulse was to pull my phone out, get chat GPT voice on and hand her the phone and say, hey, uh, why don't you talk to this? See if this thing can help you. So she did. And, um, I know, explain the situation to it. It gave her very good advice. Basically, you know, as you would expect, right? Like, I don't think you should click on that. That sounds like a scare tactic. Are they asking for money? Yeah, they, it's going to be $2 and then it's after that it's $12 a month. And, you know, the, so the AI is like, um, by the way, you know, did you give them any money already? You know, and she's like,

• 1:14:07Nathan Labenz?

  no, no, I haven't paid for anything. Okay, good. You know, close it out. Don't worry about it. Your iPhone has great security on it already. Uh, you know, as long as you don't give people your stuff, like you should be okay. So she took that advice. Okay, great.

• 1:14:21Nathan Labenz?

  Then I was trying, would you like to have this on your phone? She said, yeah, that sounds great. That was really helpful. Uh, couldn't remember her password for the iPhone. So we couldn't install the app, but I was able to just install the 1-800 chat GPT, uh, not install, just say that as a contact. Uh, and I do think that's a pretty cool service that OpenAI offers. I mean, it's, you, the initial thought for many, of course, who are into this stuff would be like, why would you need a 1-800 number? Like, who's ever going to use that? And the answer is my elderly neighbor who cannot, um, you know, cannot even remember her own password to install a new app on the phone. So she's got that now. She, you know, can call that. And it just left me with this feeling, which I've,

• 1:15:04Nathan Labenz?

  you know, approached a couple of times from a couple of different angles. But again, really drove home. There is just an unbelievable amount of adoption work to be done with the current models. This didn't push her anywhere, you know, this, this did not, um, push the model's capability anywhere close to its limits, right? This is like super easy mode for the model, but it's the kind of thing that isn't going to happen without like people actually going out and trying to help the people in their lives adopt this kind of thing. I, I, I think you're right that like the capital cycle is, does need to see results

• 1:15:45Nathan Labenz?

  soon. But I think we can get 4% growth without

• 1:15:52Nathan Labenz?

  additional hyperscaling at this point. You know, I want those medical advances, all that good stuff too. But like, do we really have to go harder on the, you know, frontier scale RL? I think the answer is no. I think that like literally just going around and helping people use what exists is enough if we actually do it. Uh, but I'm the first person who stopped by my neighbor's house and, you know, we don't have a national program like this. This is something I think would be an incredible good for the individuals who would get the help and also, you know, just to make sure that we actually have the sort of economic cushion to, you know, be able to pace without like the, you know,

• 1:16:33Nathan Labenz?

  the capital market sort of Damocles immediately coming down on us. Um, if we just put some shoulder into adoption, I think we could create a little space and would love to see more of that type of thing happening.

• 1:16:48Nathan Labenz?

  Part eight, keep the steering wheels. First, the view from inside a self-driving car.

• 1:16:56Nathan Labenz?

  Oh, I spent most of yesterday in a Tesla full self-driving vehicle, going to Chicago and back from Detroit. And, um, top takeaway from my experience yesterday, honestly, was the FSD is just so awesome. I mean, I was able to take, uh, an engagement on a day trip basis to go to Chicago, uh, show up fresh and relaxed after the thing did all the work to get me there. Then I, you know, stayed and had dinner and a drink with the, uh, the gang that I was doing the workshop with, left to come home, you know, later than I definitely would have felt comfortable driving five hours home after a big day like

• 1:17:38Nathan Labenz?

  that. And just basically reclined the, um, you got to keep your eyes open, right? It will not, uh, will not let you outright sleep, but it's so good. I just kind of reclined the seat and, um, you know, listened to my podcast, caught up on a couple of feeds on the way home. And it was just like, man, this is, I've talked so many times about how a lot of AI use cases can be GDP destroying. This one was clearly one of those that's GDP enhancing. I just could not have taken this opportunity otherwise.

• 1:18:13Nathan Labenz?

  That's the enthusiast in me. Two days earlier, I had worried out loud that we are letting our own ability to take back control atrophy and that the real doomer scenario is everything failing at once. Prakash pushed back.

• 1:18:28Speaker 04

  Maybe the doomers are being a little bit naive, though. Maybe because maybe they're expecting people to give up control. And I don't think people are going to give up control. If you look at like AI doomerism or AI safetyism, the question that I ask is, when do you think we will allow humans to not be used to fly passenger planes? Because it's been possible for 15 years and we haven't. Like, why? You have to answer that question before you say that we're going to give up our rights and we're going to be disempowered. If you can't answer that question, then you don't have a working theory of why human beings like accept technology into their lives.

• 1:19:11Speaker 04

  You're just kind of being naive about like capability kind of, you know, being there and that humans will automatically kind of adopt.

• 1:19:20Nathan Labenz?

  Yeah, I mean, I think that's that's an optimistic note and I hope that's right. I maybe, you know, maybe it's a little bit of projection by somebody like me, who is, you know, trying to live on the edge of deployment and is sometimes feeling my muscle, my personal, you know, muscles, for lack of a better term, atrophy here and there. I also do think, you know, Tesla is now making cars with no steering wheel, right? So this would maybe be a leading indicator. Like, do we start selling cars with no steering wheel and no pedals? I don't quite get why we would want to do that, honestly. Like, it's maybe it's a cool look or maybe they should be like stowable

• 1:20:05Nathan Labenz?

  in the future or something like that. But it does seem as like a metaphor for society as a whole. We probably want to keep our steering wheels in the cars versus rushing to a future. And I want the AI to do most of driving for me. But like, why are we taking it out? Why are we rushing to take it out? You know, in a sense, I still think we'd be wise to just keep steering wheels. You know, I think it's a little extra cost, maybe a little, you know, little extra difficulty in, you know, various ways, but like, probably worth keeping.

• 1:20:43Speaker 04

  I think, I think there's, number one, we, again, we have this like, how injurious it is and how easy it is like this, this like curve, right? And I think the fact is that a, an automobile accident is only injurious like one or two people. And so it's not like a plane. If you want to deploy them more widely, they should be cheaper. And you want them to be cheaper, you have this object into the steering wheel, which is not being used 99.9% of the time. So, you know, let's get rid of the steering wheel. And then if we get rid of the steering wheel and get rid of a bunch of stuff in there, we can bring down the price again, like 50 or 60%. And you bring down the price 50 or 60%, then you can expand the number of people that you can deploy to. And if we expand the number we can deploy to, then you're, you know,

• 1:21:25Speaker 04

  the number of people dying from auto crashes due to humans drops by from 30,000 to maybe, you know, 5,000, right? So I think there's this sequence of activities that you start to look at. And then that's why we go, go ahead with it. So I think it's a, it's a gradual process, like, you know, gradual human disempowerment, so to speak. So I can see why the steering wheels are being taken away. I think, I think for maybe for larger things like planes, we won't. But maybe for smaller things like, like cars, I, you know, maybe we do. I don't know.

• 1:21:58Nathan Labenz?

  Yeah, well, it does seem, I think things are kind of trending toward the, this is sort of an Eli Azar type story where like, how did you go bankrupt slowly at first and then all at once? Like, that I think is looking more and more like the story of AI takeover or human disempowerment to me all the time. It does seem like we're, I mean, we may not, but I think there's a pretty good chance that we will manage a lot of the things in the short term. And then we will be kind of rolling the dice on like, will the AIs kind of break bad at some point in the future?

• 1:22:41Nathan Labenz?

  Because we just don't have fundamental alignment science. We do have a lot of like effective control measures and better sandboxing and, you know, de-chitified RL environments. But we just won't really know, even as all that's happening, if there's like, you know, if the big one is still kind of waiting right under the surface.

• 1:23:06Nathan Labenz?

  That's the week. Thanks for listening to the AI in the AM Weekly Highlights. The narration was my cloned voice. Please tell us what worked and what didn't. See you in the morning.

• 1:25:07Nathan Labenz?

  If you're finding value in the show, we'd appreciate it if you'd take a moment to share it with friends, post online, write a review on Apple Podcasts or Spotify, or just leave us a comment on YouTube. Of course, we always welcome your feedback, guest and topic suggestions, and sponsorship inquiries, either via our website, cognitiverevolution.ai, or by DMing me on your favorite social network. The Cognitive Revolution is part of the Turpentine Network, a network of podcasts, which is now part of A16Z, where experts talk technology, business, economics, geopolitics, culture, and more. We're produced by AI Podcasting. If you're looking for podcast production help for everything from the moment you stop recording to the moment your audience starts listening, check them out and see my endorsement at ai podcast dot I N G.

• 1:25:51Nathan Labenz?

  And thank you to everyone who listens for being part of the Cognitive Revolution.