Machine Learning Street Talk / episode intelligence
Speakers in the public record
Claim mix
Evidence policy
Every row below preserves an exact excerpt. Identified speakers are linked; unresolved voices are labeled and excluded from people counts.
Claim ledger
65 published records
01 / recommendation
“I see a set of problems that a cognitive creature must solve, many of which I wrote in my 2,001 book, The Algebraic Mind.”
02 / recommendation
“And it was in a book that I wrote called The Future of the Brain, which I guess we wrote in 02/2015.”
03 / belief
“The reason this is and for fluid intelligence, such as what we see in the Arc AGI thing and Raven's regressive Matrices test, that still seems very deficient as well. So I think when thinking about these it's important to split up one's notion of cognition because if there is a severe limitation on any of these dimensions, then the models will be fairly defective economically, or at least for many tasks.”
04 / belief
“Think that's exactly what people like Sam and Dario are now hinting that they're gonna be able to do soon. I'm not really buying those claims, but I think that's what they wanna do now.”
05 / belief
“I think you have to have the background context in both cases be that you're able to proceed with development under some risk tolerance that's much lower than what there is today.”
06 / belief
“I hadn't played this variation either before, but it was obvious what to do. So it failed to identify what 3 in a row was, which I would say goes to a conceptual weakness like you should have had enough data to understand what 3 in a row was.”
07 / commitment
“You suggested 1 red line, there are others. So we could decide as a society there are certain red lines, maybe recursive self improvement might be a reasonable 1 to consider, that we could say we won't cross any of the red lines.”
08 / belief
“I think your forecasts don't make a lot of contact with the kind of stuff that the 2 of us just talked about.”
09 / commitment
“I think that we will need some Einstein level innovations in order to get to AGI and certainly to get to ASI.”
10 / belief
“I think basically both positions are legitimate, but I think if you'd integrate them, there could be a bullish case that you'll be able to knock off some of this cognitive abilities by the end of the decade.”
11 / belief
“I think with the current paradigm, I am expecting them to continue to leapfrog each other.”
12 / belief
“I agree. So, Amy, we can agree that, like, voluntary self regulation is probably not enough to get to that point.”
13 / belief
“Read I think more people probably read the, darker scenario than the positive scenario.”
14 / belief
“I mean, here's way to put it, is the AGI that I think we're afraid might be unconstrained or whatever, there really shouldn't be a lot of edge cases like that.”
15 / belief
“I think we are completely agreed that we're not doing a great job on the alignment problem and that we need to do much better and that there's a temporal dimension to that as you were just saying, which is like, you know, it's not great for humanity if we solve that problem in 200 years and we we have AGI or ASI, you know, in the next decade or 2.”
16 / disagreement
“I mean, also just, for the record, I disagree with you, I think that progress is being made in this sort of hard to measure dimension that you're pointing to.”
17 / belief
“Was going say the last thing, which is I first noted the problem with chess I think 2 years ago, the illegal move problem.”
18 / belief
“There's some differentiation in how people do their RL and what their data sets are, but I think there's not so much value in transparency until somebody has something that really is unique that they think other people aren't going to just reconstruct very rapidly.”
19 / uncertainty
“The coding tasks, I'm very concerned from a scientific perspective that we don't know how much data contamination there is.”
20 / belief
“I I think that that requires an understanding of the problem, what humans want solved by the problem.”
21 / belief
“I I think distribute I'll put it this way. There'll be many, many examples I can come up with distribution shift.”
22 / belief
“I think that you'll probably still be able to come up with some things like this a year from now, but it'll be harder.”
23 / belief
“Think that I mean, there's probably some of that there to some extent, for sure. But at least my experience at OpenAI was that there was more pressure to not talk about the risks in public and to downplay that sort of thing.”
24 / belief
“At least I would say to the same extent that humans can do things without having trained on it.”
25 / uncertainty
“I was told, I don't know if it's true, that that people in OpenAI actually look on social media for example, including mine, and, you know, people patch them up.”
26 / belief
“I know they've been trained on Wikipedia. They've probably read like Bobby Fischer played chess, which is how I learned to play chess.”
27 / commitment
“I find it very hard to know when there will be a state change. But I think in 02/1935, we will look at LLMs and be like, nice try.”
28 / commitment
“I think by the end of the year, we will already be less far ahead on agents, as you hypothesized there.”
29 / belief
“If you can have the neural networks reliably call the tools that they want or that they should be calling relative to the problem, I should say, then you're golden. I think empirically it is a bit hard to get the tools to work reliably.”
30 / commitment
“I think that the type of thing that I'm probably going to end up advocating for is going to be more of a rather than, like, here's a line that we're all not going to cross, something more like, we are going to gradually develop AIs with these capabilities, but we're going to do it in a way that's, like, mutually transparent to each other and that proceeds sort of slowly and cautiously where we all debate whether it's safe to go to the next level.”
31 / belief
“I mean, would say, in some sense, arguably, we're already seeing a move away from LLMs with things like these so called reasoning agents that have access to tools and can write code and so forth. So there's going to be this continuous shift towards, I would say, rather than, like, discrete paradigm shifts, it'll be more like a continuous paradigm shift.”
32 / belief
“I think pushing for things like transparency as well as the government having people whose job it is to be keeping track of these, coming up with contingency plans, interviewing these labs for their plans and coming up with an internal assessment, their probability of success.”
33 / belief
“I mean, I think someone could come up with a scheme to get LLMs to to play chess without making illegal moves by calling a tool.”
34 / belief
“I think the 3 of us are maybe not known for agreeing on everything, but we actually have a lot of shared values around that, and I'm looking forward to the conversation.”
35 / belief
“I agree, we're totally not on track to have figured out the alignment stuff in time.”
36 / belief
“I think that the level of intelligence that you attribute to the machines towards the end of the essay I remember some of that actually happens after the year 2027.”
37 / belief
“You know? But That's part of my darkness about it is I think there's too much alchemy and not enough principles.”
38 / belief
“You know, there's there's a lot in this in this deep future that 1 could pull at, but I I I think there are some there there are some, paths, where things work out well.”
39 / uncertainty
“Like they've done lots of And whereas 3, we don't know the details of three's training process, but it's possible that there was not a single playing of a game of chess at all in the training process for early.”
40 / belief
“I I I think making time for technical research, I'm not expecting much return on investment from that.”
41 / belief
“Like in the very fastest case, it'd be 2,030. It would require solving multiple problems that I think we're just not positioned quite yet.”
42 / belief
“Don't know about the second 2 the other 2, but definitely the first 1. I think that if somehow we could coordinate on that, that'd be great.”
43 / belief
“You can argue about that and and, you know, I I would say that like the weakest AGI would be it's as good as Joe Sixpack who's really not very good at reasoning, has is full of confirmation bias, has not gone to graduate school, doesn't have, you know, critical reasoning.”
44 / belief
“Main progress over the past 2 years or so has been in mathematical ability and short term memory is also better. But so I'm still wrapping my head around that since I'm being more noncommittal about some of these different forecasts where I think maybe it still seems very plausible, more than plausible by 2,030 something that has the cognitive abilities of a typical human, some system like that.”
45 / belief
“Like I believe in my political party and so when my political party does dumb thing then I go and try to, you know, come up with a rationale for it.”
46 / belief
“I think that that's, that's a direction or an interpretation of those facts that that had that is reasonable, but but there's there's a question of incentives and what you know, there might be a a somewhat different ask that's better.”
47 / belief
“Claude plus code interpreter and things like that. And I think that that system could solve tariff and things like that because Claude might be smart enough to look up the algorithm and then implement the algorithm and then do it.”
48 / belief
“The thing that might decrease it is propaganda from the tech companies, I think.”
49 / belief
“I think the notion of a durable advantage in LLMs, if that's what the technology is, is a myth.”
50 / commitment
“I am sure that I will be able to come up with variations on chess, that are not orthodox variations, but maybe there's some already known.”
51 / belief
“We're working on it. I agree with the problem you're pointing out. We currently have a project to make a good ending, so to speak, at a similar level of detail to what we already have.”
52 / belief
“I think AGI doesn't have a clear definition, so it's difficult for me to say for all definitions of that that would be worth standing in front of.”
53 / commitment
“I didn't see it myself, that at some point he was saying, we won't build frontier models because of these risks.”
54 / belief
“I think you could set a society up so that people have the autonomy to choose between different ways that they would want to live their life given the resources that AI increasing GDP could provide.”
55 / belief
“I think if you rule out data contamination, the performance on problems that are new is not good.”
56 / prediction
“I mean, like, if you're worried about superintelligence, then you might think if we move even towards AGI, which presumably is closer in time, that that's a bad idea because we won't be able to stop the superintelligence.”
57 / prediction
“That's And I expect that this game will keep continuing, and we won't get to a state where that is basically mostly managed in time, because the risk surface will keep evolving with agents that will present new things, we'll have to deal with those current cases that will create a substantial backlog, and we just won't have the adaptive capacity.”
58 / commitment
“I think that eventually we want to build AI because it is possible to get it right.”
59 / evaluation
“The part that I think is is more robust and more worth using is this core idea that you can can think of this trade off between more time to come up with new ideas and do AI research and, more compute with which to do the AI research.”
60 / evaluation
“We have slightly different ideas about world models. But he would say, I don't think we're close because we don't really have world models and neither of us I think are satisfied with the current thing that some people call reasoning but neither of us think is robust enough.”
61 / prediction
“I think we're making some bad assumptions, but it's hard to predict when people will move past an EDA fix, which is where I think we are now.”
62 / evaluation
“On Tower of Hanoi, Herb Simon solved it in 1957 with a classical technique that generalizes to arbitrary length whereas the LLMs do not generalize to arbitrary length and face problems. And so I could go through more but the gist of it is I don't see the qualitative problems that I think need to be solved.”
63 / prediction
“Then we have the whole I don't know if we have time to go into it, but maybe we will, maybe won't the mechanisms of open sourcing or at least open weight models, which means that essentially anybody can get in this game to some degree.”
64 / evaluation
“It works better for things like math where you can verify that the the augmented data are better. And so I think we're already running against that, kind of like bottleneck on 1 of the, let's say, raw resources that go in at least into the current approaches.”
65 / evaluation
“I take I take even, like, don't make illegal moves in chess to be a form of the alignment problem, like a very simple microcosm of the alignment problem, and they're still struggling with that.”