00:19:26 路 John Coogan
That's where the real dollars are too. The timeline is unhappy because the idea of democratizing science, technology, all of this is very alluring. But the pool of dollars available from all the biohackers in the world probably isn't close to the budgets available from big pharma. And so you, again, you're in this like rational business analyst situation and you fail to see how this is that damaging, except to like the hacker community. The real tricky part is how AI Frontier AI research is handled. Instead of outright rejecting the query and bumping the user down to Opus, the model appears to answer but quietly gives a degraded answer. And this was disclosed in the model card which is interesting. So this is again reasonable disclosed product. While they're paying for it. Which is a different which is a different path than Yeah. Bio and cyber. So if you if you go LLM frontier research devil horns, it will actually give you an answer apparently. It won't bump you down immediately. But so it doesn't disclose that, which is odd. Outright rejecting requests for AI research and just saying, hey, user, sorry this model doesn't work for that type project. Please share please use another model or contact sales if you want help with this, would have been much more in line with the bio and security and cybersecurity strategies. And they and then they also could not have disclosed and it's also possible that they just didn't need to disclose this at all. Like, they could have just released a model that was intentionally nerfed on AI research. It would have shown up in the benchmarks because people would have benchmarked it on some sort of AI research bench and been like, oh, weird. It's really good at all these other things but it's bad at LLM research. And maybe that would have been a bit of a brand hit maybe, but users might never know that the model was intentionally degraded around this category of work. So that leaves this third more worrying position, intentional degradation without disclosure in the model card. There's no evidence of this, but it's possible that other workflows might be nerfed and there's no law or even convention around disclosure. Again, maybe good business, but a weird situation to be in. So probably bullish for evals if you're building a business on top of a big lab.
00:21:33 路 John Coogan
You can imagine like a legal AI company who want to be really sure that the models they're using aren't degrading unexpectedly and not telling them. It's different if you're like, hey, I've been a bio researcher for a while. I'm using this and I know that this model was never intended for me. Or or I've been using it, but what you don't want is like I'm using it and now it's giving, now it's leading me astray in my work and it's also not telling me that it's going to lead me astray, which would be like sort of an odd outcome that I think they'll probably address in the near future.
00:22:05 路 Jordi Hays, John Coogan
[00:22:05] S0: Anyway. Yeah. That aspect triggered Dean Ball. What He he said, My last observation re anthropic secret sabotage safety policy is that it undermines actually good safety policy. How? First, it is very plausible to describe this as anti competitive behavior. Even if you're maximally sympathetic to Anthropic here, you must admit this and it behavior being justified in the name of AI safety. If you believe, as I and many Anthropic staff do, that it may end up being critically important to relax antitrust enforcement so that the Frontier Labs can cooperate and collaborate on some areas of AI safety, Anthropic just undermined the case for that in a large way. Overall, this massively and profoundly raises the status of the argument that AI safety has been hyped to justify monopolistic behavior by labs. I continue to believe that AI safety is a real and serious issue that is growing in importance rather than diminishing. If you agree with me, this incident is a setback, maybe a serious one. And third, says, as I have observed elsewhere, Anthropics official corporate policy is structurally identical to the fact pattern alleged against them by the Department of War. I still think DOW acted both falsely and wrongly in that fight, but it is no longer possible to defend Anthropic with a full throat after this incident. This raises the case for heavier handed regulations. Anthropic is making awfully good case here that their product ought to be treated as utilities and thus their alignment practices should [00:23:32] S1: Oh, be a matter counter carrier thing.
00:23:32 路 John Coogan, Jordi Hays
[00:23:32] S1: Oh, be a matter counter carrier thing. [00:23:33] S0: Public policy rather than private property. I am starkly opposed to this sort of state power grab but Anthropic is doing more to justify it than anyone else. Thus significant damage has been done to community and entire approach to AI governance. It was done unilaterally by Anthropic likely motivated largely by self interest and justified within the internal psychology of the firm through the lens of safety. Suspect this is fixable in the economic and legal senses for Anthropic, but I fear that trust has just been broken and the goodwill extinguished will take very much time to repair.
00:24:04 路 John Coogan
[00:24:04] S1: And just to level set on Deembal, he wrote the AI action plan, but then also came out very publicly in support of Anthropic during the conflict with the Department of War, saying that the Department of War is completely overstepping by pushing towards supply chain risk designation, putting pressure on Anthropic for not wanting to work with the government in that particular way. Obviously there's a whole bunch of new data that's been released and stuff and that conversation has evolved. But he's not is he I he doesn't strike me as some like crazy hater. Anyway, let me tell you about Railway. Railway is the all in one intelligent cloud provider, user favorite agent to deploy web app servers databases and more while Railway automatically takes care of scaling monitoring and security. Doug O'Laughlin was also sort of mixed results on Fable. He says, when it works, it's brilliant, but the unilateral guardrails make me frustrated beyond all belief. He has a folder of health information with like a hundred days of Aura Health data. This is a very logical thing that people would do with Codex or Cloud Code, gather up all your health data. I actually talked to a very, very prominent person in tech about how they got into vibe coding and CLIs. And one of the first things they did was reanalyze some WHOOP data, reanalyze some health data and he correctly detected sleep apnea and went and got it treated. Like that is it's not the cure for cancer but that's like incredibly awesome and like very, very good and exactly what you want to see. And who knows, maybe the thing that you detect early could lead to an increased risk of cancer in the future. And that would be a really, really big win for our society. And so a little bit tricky there. He says there's a private investment in the life sciences tool space. Guess what? It's not safe. Doing some code scanning for vulnerabilities, not safe. I get the safety but it feels incredibly out of touch for a group of a few 100 a few thousand people making all making total comp in the millions, telling me what is and isn't safe. Dario worried about inequality, I think he has to realize that he is the inequality and the unilateral gatekeeping feels whack as hell. I don't like it. Yeah.
00:26:16 路 Jordi Hays, John Coogan
[00:26:16] S0: People are people are very [00:26:18] S1: And like Doug O'Laughlin is extremely optimistic about Anthropic. Like he he he recognizes the power of the models in the business and has been a very very strong supporter. I I feel like a strong user. He was one of the first people to admit to Claude code psychosis and he's unhappy with the current situation. But these things are very tunable. I'm optimistic with more disclosures and more fine tuning and and a and a smoother smoother path. I think we can get to the good outcome here. Cremieux. Last one, [00:26:51] S0: Cremieux says, tell me about It's the powerhouse of the cell. Right? It's chat The [00:26:58] S1: chat pause is rough. Yeah. It's a it's a screenshot. [00:27:02] S0: But maybe maybe the harness is just saying like, sorry, if you don't know that the mitochondria is a powerhouse of the cell like [00:27:10] S1: It's not worth my time. I don't even want your I mean, that's the thing with like the good [00:27:14] S0: You know, there are dumb questions.