ClaimsPeople

Eliezer Yudkowsky

Eliezer Yudkowsky founded the Machine Intelligence Research Institute (MIRI) and LessWrong, co-authored the best-selling book 'If Anyone Builds It, Everyone Dies' (Little, Brown, 2025), and was named to the TIME100 AI 2023 list.

2,016 claims collected, 236 researched, 79 resolved. Report of 4 November 2026, method 0.1.0+c876030c.

Scores by kind of claim

Each kind of claim is scored on its own, out of 100: the share that held up, with the range it likely falls in. Fewer resolved claims give a wider range.

  • Factual claims

    Statements about the present or the past, checked against the record: did they hold up?

    76 70 82

    High confidence, 71 factual claims resolved. From 16 documents, 2007 to 2026; 18 documents sampled from 23 candidates found.

    Not scored: 50 unresolvable, 8 checkable only against their own records.

  • Predictions

    Statements about events outside the person's control, checked once they fell due: did they come true?

    60 42 78

    Low confidence, 8 predictions resolved. From 4 documents, 2008 to 2024; 18 documents sampled from 23 candidates found.

    Not scored: 97 unresolvable. 1 more fall due later.

    Does not yet allow for how likely each prediction was when made.

All types combined: 74 (likely 68 to 80, high confidence), from 79 claims: 71 factual claims, 8 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.

Eliezer Yudkowsky's claims

2,016 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Not checkable Not checkable 6 Apr 2023 To me it just feels like patterns in the clouds and maybe that cloud actually is shaped like a horse. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The concrete stuff that is safe, that cannot kill you, does not exhibit the same phenomena as the things that can kill you. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 As we ran out of time to improve civilization. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The person on this planet who is doing the alignment work thought those kinds of thoughts and I am skeptical that it decouples. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Getting stuff to be more likely does not help you if the baseline is nearly zero. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Natural selection regularizes so much harder than gradient descent in that way. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The prediction is when you give yourself more options, you start to select ones that look weirder relative to the ancestral distribution. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 AI is nuclear weapons but they spit up gold until they get too large and then ignite the atmosphere. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 They are not evidence one way or the other because the basic prediction is like, if you offer things enough options, they will go out of… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I don’t think the sequences were good writing by my current standards but they were good enough. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I think I’d have to explain that poll pretty carefully because they haven’t got the intelligence headbands yet. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Yeah, it’s kind of weird because small, large amounts of intelligence don’t automatically make you a computer programmer. Not checkable · 6 Apr 2023
Unresolvable Prediction 6 Apr 2023 Yudkowsky said he does not think we will live in the "sane world" that would "shut it all down and work on human intelligence enhancement", and that he thinks we are all going to die. Prediction · 6 Apr 2023 AI safety and alignment
Not checkable Not checkable 6 Apr 2023 Humans came out pretty different from their loss functions. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The super vast majority of predictions of utility functions are incompatible with humans existing. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 First of all, even if you have something that desires no power for its own sake, if it desires anything else it needs power to get there. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The main thing is that you’re starting from minds that are already very, very similar to yours. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The tiny shred of hope, which I tried to jump on with the Time article, is that maybe people can understand this on the level of — “Oh, you… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Dive under the surface, look at the things that have changed. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Okay. Whoa, whoa, whoa. Flourish? Keeping in a zoo did not sound like flourishing to me. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 That would be a pleasant fantasy for people who cannot abide the notion that history depends on small little changes Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Our society does not have good words for these things. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 If the world of AI had looked like way more powerful versions of the kind of stuff that was around in 2001 when I was getting into this… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 None of this is very analogous to AI. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Yeah, I think my concern is that that works better when the things you’re breeding are stupider than you as opposed to when they are… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Man, I feel like many of those ships have already sailed Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I guess because I could be wrong and because matters are now serious enough that I have nothing left to do but go out there and tell people… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I think that the Hanson-Yudkowsky foom debate was won by Gwern Branwen Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 My model no doubt has many errors. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 If the humans are no longer around, you no longer need to predict them. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 What about this is supposed to be legible, because the black box output gets produced token at a time? Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 This seems like an ordinate cope. Not checkable · 6 Apr 2023
Mostly true Factual 6 Apr 2023 Yudkowsky previously thought that building intelligence would involve multiple specialized systems. Factual · 6 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 As of April 2023, AI developers are not being careful and deliberate in designing AI systems. Factual · 6 Apr 2023 AI safety and alignment
Not checkable Not checkable 6 Apr 2023 No, we’re not saving the information. We’re doing a total rewrite to the DNA. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Eliezer’s get more dangerous as you lock them in a room with aliens they do not like instead of with humans Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Your skepticism disproves too much, my friend. Not checkable · 6 Apr 2023
Unresolvable Prediction 6 Apr 2023 At some point AI systems will become smart enough to build their own AI systems better than humans can. Prediction · 6 Apr 2023 AI capabilities
Unresolvable Prediction 6 Apr 2023 Yudkowsky said he was not over 50% confident that "they" would actually carry out the "shut it all down" option. Prediction · 6 Apr 2023 AI policy and governance
Resolved true Prediction 6 Apr 2023 AI alignment will not be solved within a few years. Prediction · 6 Apr 2023 AI safety and alignment
Not researched Factual 6 Apr 2023 AI algorithms are continuing to improve. Factual · 6 Apr 2023 AI capabilities
Not checkable Not checkable 6 Apr 2023 The drama is meaningless. What changes the chance of victory is meaningful. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 You are not training a human there. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 It’s being trained on Napoleon’s words. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 GPT-4 was just suddenly acquiring this new batch of qualitative capabilities compared to GPT 3.5. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I claim that I am honest. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 It might possibly have to find a security flaw somewhere on the AWS or Azure servers running it. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 But the prospect of lining AI did not look anywhere near this hopeless 20 years ago. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Sure, lots of instrumental values made their way into us but even more strange, warped versions of them make their way into our intrinsic… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The point is, reasoning off of humans is pretty hard. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 At the core I think what you’re describing there is a sufficiently functional society and civilization that could understand that — if they… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I just don’t trust the blind hope that all of that capability is pointed entirely at pretending to be Eliezer Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 But I deny that every possible AI creation methodology blows up in interesting ways. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I claim humans are increasingly orthogonal and the further they go out of distribution and the smarter they get, the more orthogonal they… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 It’s all about the ignorance prior. It’s all about knowing the space in which to be maximum entropy. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I would reserve the term moral updates for humans. Not checkable · 6 Apr 2023
Unresolvable Prediction 6 Apr 2023 Yudkowsky said he would expect that, even if a full AI shutdown were adopted, those responsible would have no exit strategy until the world ended, though he added that perhaps he underestimates them. Prediction · 6 Apr 2023 AI policy and governance
Not checkable Not checkable 6 Apr 2023 We’re currently nowhere near as far as we could be. This is not the glorious transhumanist future. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 But it’s not the average. It is able to be every one of those people. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 these things don’t have names. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 That means that somewhere in the giant inscrutable vectors of floating point numbers, there is the ability to plan because it is predicting… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 If we were on a sane planet, what the sane planet does at this point is shut it all down and work on human intelligence enhancement. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I am not very high on us doing it. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 That’s what it looks like to be a Bayesian. Not checkable · 6 Apr 2023
Resolved true Prediction 6 Apr 2023 GPT-5 will not end the world, with more than 50% of Yudkowsky's probability on that outcome. Prediction · 6 Apr 2023 AI safety and alignment
Not checkable Not checkable 6 Apr 2023 I’m not sure it’s enough to not have Hitler, but it sure would be a start on things going differently in a timeline. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 The universe with the most predictable text is not a universe that has humans in it. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Why are people asking for the most difficult thing that’s the most impossible to verify? It’s whack. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I feel like we bit off a whole chunk of chimp to human in GPT 3.5 to GPT-4, but go on. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Okay but then you’re claiming things as evidence for your position. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 It will lack dignity not to try. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 They’re complicated. There’s not a short list. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Because it was in the interest of neither party to have a full nuclear exchange. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 And you can’t calculate the exact point at which they ignite the atmosphere. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I watch other people be fooled by stuff that would not fool me. Not checkable · 6 Apr 2023
Unresolvable Prediction 6 Apr 2023 If an AI is made smarter until Yudkowsky starts approving its outputs, it will find a way to fool him. Prediction · 6 Apr 2023 AI safety and alignment
Not checkable Not checkable 6 Apr 2023 They don’t know any better. Not checkable · 6 Apr 2023
Unresolvable Prediction 6 Apr 2023 If a very smart AI system is trained under the present paradigm, it will kill us before it can be tuned with RLHF. Prediction · 6 Apr 2023 AI safety and alignment
Not checkable Not checkable 6 Apr 2023 And it’s going to be more dangerous than what I was dreaming about in 2003. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I expect things to continue on largely as they have. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 You do not just turn into a random human because the random human is not what’s best at predicting the next word of everyone who’s ever… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 It’s a genie, it gives you what you wish for. Not checkable · 6 Apr 2023
Not researched Factual 6 Apr 2023 Eliezer Yudkowsky said he tried very hard to replace himself. Factual · 6 Apr 2023 Field building and careers
Not checkable Not checkable 6 Apr 2023 I don’t know how to throw a billion dollars at outputting lots and lots of alignment stuff. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 And the long term, we sure aren’t. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 If you still want to, come the day, I don’t think I myself would oppose it. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Yeah, that’s not orthogonality. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 And they are not really here. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 And there’s a whole bunch of caution to be had about not making them smarter and making them evil at the same time. And yet I think that… Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I am specialized on alignment rather than persuading humans, though I am more persuasive in some ways than your typical average human. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 In 2008, I did not know this stuff was going to go down in 2023. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Where do people have this notion of getting AIs to help you do your AI alignment homework? Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 if you take a human and make them smarter, that may shift their morality. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I feel like this is not likely to be truly very informative by my standards Not checkable · 6 Apr 2023
Unresolvable Prediction 6 Apr 2023 Yudkowsky thinks the approach will predictably break down as systems are made smarter and as one tries to get sufficiently useful work from them. Prediction · 6 Apr 2023 AI safety and alignment
Not checkable Not checkable 6 Apr 2023 Stop asking for help with alignment. Ask it for any of the help. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 I’m not going to give clever details for how it could do that super duper effectively. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 It’s more of a Hail Mary pass. Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Whoa. Where do you get a good world out of maximum prediction of text? Not checkable · 6 Apr 2023
Not checkable Not checkable 6 Apr 2023 Maybe we don’t want to conserve the parts of nature where things bite other things and inject venom into them and the victims die in… Not checkable · 6 Apr 2023
Back to the list

Not checkable

“You do not just turn into a random human because the random human is not what’s best at predicting the next word of everyone who’s ever been on the internet.”

Eliezer Yudkowsky 6 April 2023 Eliezer Yudkowsky - Why AI Will Kill Us, Aligning LLMs, Nature of Intelligence, SciFi, & Rationality (Dwarkesh Podcast)

You do not just turn into a random human because the random human is not what’s best at predicting the next word of everyone who’s ever…

Not checkable An opinion, value judgement or statement with nothing to check.

Cite this claim

Eliezer Yudkowsky, “You do not just turn into a random human because the random human is not what’s best at predicting the next word of everyone who’s ever…” (said 6 April 2023). Verdict: not checkable. Reliability API, claim c_a934528592bb, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_a934528592bb

Claims without a verdict

Is a verdict here wrong, or do you speak for Eliezer Yudkowsky? Reply on the claim's own page, with the form at its foot (how replies work). Verdicts are made automatically; how verdicts and scores are made.