ClaimsPeople

Dario Amodei

Dario Amodei is co-founder and CEO of Anthropic and a former VP of Research at OpenAI, where he led work on GPT-2 and GPT-3.

1,335 claims collected, 337 researched, 97 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 69 82

    High confidence, 76 factual claims resolved. From 14 documents, 2015 to 2026; 19 documents sampled from 30 candidates found.

    Not scored: 51 unresolvable, 42 checkable only against their own records.

  • Predictions

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

    56 42 70

    Medium confidence, 17 predictions resolved. From 8 documents, 2015 to 2026; 19 documents sampled from 30 candidates found.

    Not scored: 103 unresolvable, 1 checkable only against their own records. 39 more fall due later.

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

  • Promises

    Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?

    Insufficient record: 4 promises resolved, 5 needed for a score.

    Not scored: 3 unresolvable, 1 not yet due.

All types combined: 72 (likely 66 to 78, high confidence), from 93 claims: 76 factual claims, 17 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.

Dario Amodei's claims

864 claims · page 3 of 9

864 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Not checkable Not checkable 5 Jun 2025 I feel the urgency personally. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 Sharing information about recursive self-improvement and about the misalignment of models can help to convince everyone that it is not in… Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 More intelligent models are more capable of deceiving tests, and thus may appear aligned while having serious problems that go undetected. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 We have tried to prioritize caution over speed and prudence over profit. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 I agree with Secretary Bessent that a Chinese lead in AI would pose grave danger for the United States and the world. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 I think such an agreement would be difficult but just on the edge of being possible. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 Left unchecked, it could outrun our ability to understand and control these systems, and so must be pursued very carefully, if at all. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 It’s easy to dismiss this incident because no one was hurt and the economic damage was minimal, but in my opinion, a swarm that possessed… Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 We have sought a middle way: to show that it’s possible to build carefully and succeed commercially, and to make safety something on which… Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 Despite all the progress, we still only understand a tiny fraction of what goes on inside these models. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 Many things go wrong not because companies are missing some important theory or insight, but because of problems in execution. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 To be clear, pacing does not mean halting model training or technical progress, but ensuring companies take adequate time to align and… Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 This is the key step for verifiability of any pacing commitments Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 I believe it’s incumbent on every frontier AI company to act as if OAI-HF had happened to them. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 The most effective method of pacing is via regulation that targets all US frontier AI companies, as that covers even those who are… Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 A race to the bottom, spurred by commercial incentives, can make these risks more acute. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 They include the risk of losing control of AI systems, misuse of AI for cyberattacks and bioterrorism, and serious economic disruption. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 The measures I propose to advance the frontier at a safe pace will not be easy. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 But like many technologies before it, AI brings risks, and because it is such a powerful technology, these risks are serious. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 We must slow the pace at which we improve the capabilities of AI models. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 Not building the technology deprives humanity of benefits or simply places AI in the hands of authoritarian powers, while building it too… Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 Carefully wielded, AI can be the latest in a long line of technological miracles that have uplifted and ennobled humanity. Not checkable · 5 Jun 2025
Not checkable Not checkable 5 Jun 2025 But I believe we owe it to humanity to try. Not checkable · 5 Jun 2025
Not checkable Not checkable 22 May 2025 the water is rising everywhere. Not checkable · 22 May 2025
Not checkable Not checkable 22 May 2025 It really depends how you measure it Not checkable · 22 May 2025
Not checkable Not checkable 10 Mar 2025 But that would not be my bet Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I have a functionalist view of, you know, moral welfare of the nature of experience, perhaps even of consciousness. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 If we’re deploying millions of them and we’re not thinking about the experience that they have, and they may not have any. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think some function that does that measurement seems very important. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 it doesn’t mean you’re convinced, but maybe you should pay some attention to it. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 it’s better if we build those datacenters in the U.S. or its allies than if we build them in countries that have divided loyalties Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 the ambition to do very difficult things Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 the last four years have actually been a kind of, you know, almost a side-by-side experiment Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 could allow an unskilled person, simply by talking to the model and following its instructions, to do things that you would have to have… Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I am actually not that pessimistic Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think awareness is step one. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think the worries on the economic side are just as existential as the worries on the national security side. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 So controlling them, making them reliable, is very difficult Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think it’s the most important thing. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think we have an extraordinary opportunity, as I said, to cure major diseases Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 They’re not like code, where you can do formal verification Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 Some of the concepts are strikingly similar. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think our commitments have held up pretty well Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 AI systems maybe can help us to do that better, but I think that will always be one of the quintessential challenges of being human. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I am struck by how meaningful activities can be even when they are not generating economic value. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I would describe myself as supportive of this dialog, but not especially optimistic that it would work. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I don’t feel like I have the answer to them. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I acknowledge that is possible. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 And we should really think about, you know, do these things have, you know, real experience that’s meaningful in some way. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 we’re not counterterrorism experts. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 more help from the U.S. government in helping to defend our companies against this risk is very important. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I actually think the most societally divisive outcome is if randomly 50 percent of the jobs are suddenly done by AI Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 And so we’ve had a very productive dialog with them on these issues. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 if it quacks like a duck and it walks like a duck, maybe it’s a duck. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 we really needed to do a good job of building it Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 human chess champions are still celebrities. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 That doesn’t mean the export controls failed Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 You know, the institute is probably misnamed. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 that actually keeps me up at night a lot, and is why I’m kind of trying to spread the message to more people. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think human productivity will actually be enhanced Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 let’s say AI causes the economic growth rate to be 10 percent a year. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 because we grow these systems we don’t train them, I don’t think it’s safe to assume they’ll do exactly what their human designers or users… Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 One thing I often say is we grow them more than we build them Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 It’s something we should think about very seriously. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 these are system-level diseases Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 And we may want to clear away some blockers. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 the export controls are actually quite essential Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I may be right or I may be wrong. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think DeepSeek was actually an example of the scaling laws Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I am concerned about industrial espionage of the companies in the U.S., companies like Anthropic. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 it should be on the radar of those people. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 the fact that I am not the world champion does not Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 we believe that understanding what is going on inside these models is a public good that benefits everyone Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 we should be worrying a lot more about smuggling these GPUs than, you know, we’re worried about smuggling, you know, guns, or even drones… Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 giving the model a button that says, “I quit this job,” that the model can press Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 It’s one of the things where we hope to kind of be leaders in defining what the appropriate use of AI for national security is. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I was surprised to learn there are surprisingly practical things you can do. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 It is a very hard question to answer. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 I think it is a big risk that humans will misuse the AI systems. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 These are the so-called reasoning models Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 So we need to do clinical trials. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 they would have huge implications for the economy Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 And so I think we should at least consider the question of, if we are building these systems and they do all kinds of things like humans as… Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 struggling through our relationships with other humans, our obligations to them Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 everyone believes that there should be limits. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 You don’t truly know what they’re capable of until they’re deployed to a million people Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 If we can solve these with AI at a baseline, regardless of kind of the job situation, we will have a much better world Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 And that actually worries me Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 But that doesn’t mean it’s not the most important thing. Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 We’re all useless is a nihilistic answer Not checkable · 10 Mar 2025
Not checkable Not checkable 10 Mar 2025 If the requirement is you have to be the best in the world at something in order for it to be somehow spiritually meaningful for you, I… Not checkable · 10 Mar 2025
Not checkable Not checkable 28 Feb 2025 It had the environment of a trade show and was very much out of spirit with the spirit of the original summit that was created in Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 Those stupid refusals are a side effect of the things that we actually care about Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 I think on the side of risks, I often feel that the advocates of risk are sometimes the worst enemies of the cause of risk. Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 On one hand, you're like, oh man, the training worked like these models are robustly good. Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 We recognize that as an oversight. Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 Probably your number one goal is don't look like a fool. Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 Yeah, so actually that is a misquote. Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 So not dangerous per se, Not checkable · 28 Feb 2025
Not checkable Not checkable 28 Feb 2025 We know how to make these. Not checkable · 28 Feb 2025
Back to the list

Dario Amodei

Open its page

Not checkable

“A race to the bottom, spurred by commercial incentives, can make these risks more acute.”

Dario Amodei 5 June 2025 We Must Pace the Frontier (darioamodei.com)

A race to the bottom, spurred by commercial incentives, can make these risks more acute.

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

Cite this claim

Dario Amodei, “A race to the bottom, spurred by commercial incentives, can make these risks more acute.” (said 5 June 2025). Verdict: not checkable. Reliability API, claim c_d594dbe6ee92, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_d594dbe6ee92

Claims without a verdict

Is a verdict here wrong, or do you speak for Dario Amodei? 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.