The Ledger, every show

“We think that, you know, AGI is within reach in the next couple of years.”

“Which is that in the next two or three years, given the right feedback loops, given the right compute, given the right, you know, elbow grease and this kind of stuff, we think that we as the AI industry are all on track to create something that is at least as …”

“I don't think that's true. I think we haven't yet really found anything that transformers haven't been able to model provided sufficient data and sufficient compute.”

“The fastest route, or like, the most immediate one is whenever we see a really good blog post where people have, like, done incredible amount of work in an independent fashion, it's one of the highest signal things there is.”

“DeepMind, if you wanted to solve science, is the best place in the world. Like, I think that DeepMind will directly contribute to more scientific discoveries from AI than anything else, right?”

“Like really for the last five or six years, Anthropics ethos has been scaling compute with broadly the current set of techniques is like AGI is tractable within those bounds.”

“We think that one of the most important signals of whether or not we are basically the speed of takeoff, the speed of progress is driven by how much AI is able to assist AI research.”

“In many ways, this model is what caused PMF for Cursor. Cursor took off like a rocket, right, with because they were in the right place, and they were able to capitalize on that model as offering a coding experience that didn't previously exist.”

“We asked it to build something that looks roughly like a chat app, you know, something like Slack or, you know. And it was, it, the model just worked for 30 hours. Like, it was just spinning there on a computer for 30 hours, and came out with a really good wor…”

“And so I think it's like every couple of months, the time horizon that the AIs are capable of doing is doubling or something, something crazy. Maybe, maybe every six months the time horizon doubles”

“Over the next six months, over the next year, expect dramatic progress here. And like look at where we are now versus where we were a year ago. And the difference is I expect the same jump basically.”

“Right now you need to intervene quite frequently, but it's usually on questions of taste rather than it is questions of, like, raw programming ability.”

“Saying, I don't know, or solving, you know, hallucinations is, ah, intrinsically requires reinforcement learning in many ways.”

“Give it math questions, tell it whether it got them right or wrong, and the model will learn. This is, it comes down to a bit of lesson in scale and search, is just allow the model to search, have enough compute to run the experiments, and the model actually e…”

“When I look at an LLM training pipeline, it is two and a half years of best effort, last minute, desperate effort. And there's just so much room to go on every part of it.”

“We'll probably reach like better than human on the GDP eval, and it won't change anything economically because It'll be all the connective tissue, and all the, like, you know, the context, and actually, like, the task won't be representative.”

“If coding agents progress in the way I've been saying, in a year or two, you'll be able to manage a team, basically, that works 24 seven for you doing work.”

“Locomotion's kind of solved, to be honest, with basic RL.”

“I really do think that similar to how we saw with large pre-trained models before with small fine-tuned models made it like, had gains over the sort of GPT-II era, but then were obsoleted by GPT-IV being generally good at everything. I think, to be honest, you…”

“And I mean, I think if you are old on Arc AGI, then it would, you'd probably get superhuman at it pretty fast. But I think we're all trying not to RL on it so that it functions as like an interesting held out.”

“So finally, this year is where the compute, like, super cycle is, like, beginning properly in effect.”

“This ended up saving several hundred million dollars, I think, like, even over the first six months”

“Anthropic has been laser focused on on two things. One is, like, long-term AI alignment, and two is near-term economic impact. So Anthropik has been laser-focused on coding and computer use, and things that we think will make a direct impact to the economy, li…”

“You know, one thing that Anthropik, like, You noticeably hasn't focused on compared to DeepMind and to OpenAI is is mathematical reasoning, right? DeepMind and OpenAI have been pursuing mathematical reasoning because of the implications for science and for sci…”

“Generations of people have developed clever methods of encoding priors about how they think an artificial intelligence should reason. And encoding it into the model, and all of this gets wiped out by scale and, you know, planning, like, basically, like, search…”

“I once asked this question of Noam Chazier. And he was like, yeah, maybe like 10% of my ideas work, and that's not, right? You know, one of the, you know, an absolute genius, one of the best in the field. So if only 10% of his ideas work, then I think that, yo…”

“A really high ROI use on, on your time would probably be to just go and look at the data and think hard about what, like the model is learning or doing and like make some tweaks that makes it like, there's so many, like you could even just The simplest things …”

“Coding is a uniquely tractable problem in some respects for the techniques that we have in terms, the data exists in many ways. You can containerize and run things in parallel. You can run unit tests, and so you can verify.”

“The Cognition folks in Devon found the model so useful, they had to, like, rebuild their architecture around it.”

“Over time you know, over the next couple of months, you're probably gonna end up in a situation where you only need to supervise the models every, you know, 10 minutes, 20 minutes or so.”

“I think one of the things that made me remarkable about the current generation of agents is that they can self-correct. In fact, they're astonishingly good at self-correcting.”

“And in this case, the model replicated Claude.ai with artifacts, with everything else I can't quite remember how long that one took. Maybe a couple hours to do.”

“Like on the meter eval, if you look at progress over the last two years, you can plot it with a straight line, right?”

“I think also an important change in, you know, in this sort of like era of reasoning models and RL on language models is, at the end of last year, RL on language models finally started to work.”

“And I think OpenAI deserves a lot of credit for you know, releasing the first, like, serious RL plus LLMs release with O-one. And I think this really kicked off a pretty, you know, substantial change because it opened up a new axis of scaling, right? There was…”

“One of the craziest things about RL on language models in the, in, like, the RL from verified rewards regime, is it's almost the simplest possible thing. It's, like, almost too simple to work. And this is, again, comes back to that question of taste, where Rea…”

“I actually think Moravec's paradox is a little bit fake, and I think this is mostly a question of data availability and like you know, RL signal and stuff.”

“We're still seeing these huge gains where you go, you know, 10 X compute increase in RL. We still getting like very distinct linear gains for that.”

“There's now this two paradigm regime where previously you did free training scaling and reinforcement learning scaling, and now we're in a mix of the two basically. And so I think that gives you more opportunities to update models because it means that you can…”

“Even if you, as much as you wanted chips last year, it would have been impossible to get them because TSMC was, you know, booked out and so forth.”

“One of the interesting things about this most recent release is actually Sonnet is smarter than Opus.”

“So that allows you to take a mid-tier model and make it as good as a larger-tier model of six months ago or three months ago.”

“Convolutional neural networks will be better than a more general, like, a vision transformer for the vast majority of images up to a certain, and up to a certain amount of scale. But past a point, actually, you need to be able to flexibly integrate information…”

“But if you look, Gemini is like, you know, a thousand people, there's a, there's still like 10,000 plus people doing all kinds of really like longterm foundational research at DMI.”

“It's not any one critical breakthrough. It's more the accumulation of a huge amount of work in an environment where there's a sort of exogenous force of compute pushing progress forward.”

“One way of thinking about this is test time compute is doing a lot of reasoning, and then RL is the feedback signal on whether or not that reasoning was right or wrong.”

“Test time compute lets you do harder problems than you can currently do, than you can like currently do off the cuff, and RL Then allows you to sort of distill that back into the model.”

“We've got very little rewards right now, but pretty quickly over the next year or two, you're going to start to see much more meaningful and long horizon rewards.”