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

157 claims · page 2 of 2

157 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Unresolvable Prediction 8 Aug 2023 AI models will undertake extended tasks. Prediction · 8 Aug 2023 AI capabilities
Unresolvable Prediction 8 Aug 2023 Dario Amodei said it seemed likely to him that AI systems will first speed up human productivity, then equal it, and at some point become in some meaningful sense the main contributor to scientific progress, though he expected the details to be different from what people expect. Prediction · 8 Aug 2023 AI capabilities
Unresolvable Prediction 8 Aug 2023 The models' current level of performance, which the speaker graded as a B or B minus, will change with scaling. Prediction · 8 Aug 2023 AI capabilities
Unresolvable Prediction 8 Aug 2023 Dario Amodei does not expect RL training for longer-horizon tasks to require a substantial amount of additional compute. Prediction · 8 Aug 2023 AI compute, hardware and data
Unresolvable Factual 8 Aug 2023 Interpretability work on language models, such as induction heads, suggests much of the cognitive machinery needed for active agents is already present in base language models. Factual · 8 Aug 2023 AI capabilities
Unresolvable Factual 8 Aug 2023 AI language models learn to speak fluent Base64, which humans do not typically learn. Factual · 8 Aug 2023 AI capabilities
Unresolvable Prediction 8 Aug 2023 Serious danger from scaled AI models will first come from misuse, such as biorisk, rather than misalignment. Prediction · 8 Aug 2023 AI safety and alignment
Unresolvable Prediction 8 Aug 2023 Dario Amodei said he personally thinks it is very unlikely that AI scaling laws will just stop. Prediction · 8 Aug 2023 AI capabilities
Unresolvable Factual 13 Oct 2022 Amodei said he does not believe US export controls were ever designed to prevent China from obtaining a few tens of thousands of chips. Factual · 13 Oct 2022 AI policy and governance
Unresolvable Factual 13 Oct 2022 DeepSeek, a Chinese company, was the first to demonstrate the expected AI cost reductions, which had never before happened with a Chinese company. Factual · 13 Oct 2022 Economics and markets
Unresolvable Prediction 13 Oct 2022 US AI companies will soon achieve similar cost reductions, through the usual cost-reduction trend rather than by copying DeepSeek. Prediction · 13 Oct 2022 Economics and markets
Unresolvable Factual 13 Oct 2022 Anthropic finds Claude 3.5 Sonnet remains much better than DeepSeek-V3 on some key tasks such as real-world coding. Factual · 13 Oct 2022 AI capabilities
Unresolvable Prediction 21 Jul 2017 Solving biology might turn out to be an exponentially complicated combinatorial problem or be limited by data and experiment. Prediction · 21 Jul 2017 Biosecurity
Unresolvable Prediction 21 Jul 2017 A really powerful AI might be unable to understand macroeconomic systems due to data limits yet still be able to design a better one. Prediction · 21 Jul 2017 AI, the economy and work
Unresolvable Prediction 21 Jul 2017 Machines could probably make progress in biology pretty fast, though there is huge uncertainty. Prediction · 21 Jul 2017 Biosecurity
Unresolvable Factual 21 Jul 2017 Vicarious does not think about AI safety, to Amodei's knowledge. Factual · 21 Jul 2017 AI safety and alignment
Unresolvable Prediction 21 Jul 2017 Amodei said he is optimistic that researchers will gain a better understanding of transparency inside neural nets. Prediction · 21 Jul 2017 AI safety and alignment
Unresolvable Prediction 21 Jul 2017 Solving resource constraints through AI will not by itself solve war. Prediction · 21 Jul 2017 Other global risks
Unresolvable Factual 21 Jul 2017 Dario Amodei mainly works on safety at OpenAI. Factual · 21 Jul 2017 AI safety and alignment
Unresolvable Factual 21 Jul 2017 The Concrete Problems in AI Safety paper did not discuss the transparency of neural nets. Factual · 21 Jul 2017 AI safety and alignment
Unresolvable Prediction 21 Jul 2017 Pure rule-based systems will not work for building AGI. Prediction · 21 Jul 2017 AI capabilities
Unresolvable Prediction 21 Jul 2017 Artificial general intelligence will eventually be developed. Prediction · 21 Jul 2017 AI timelines and forecasting
Unresolvable Prediction 21 Jul 2017 There may be more than a 50% chance that, as AI gets closer to AGI, it becomes clear how to make it safe. Prediction · 21 Jul 2017 AI safety and alignment
Unresolvable Factual 21 Jul 2017 Statistical AI systems have pretty decisively won out over rule-based systems. Factual · 21 Jul 2017 AI capabilities
Unresolvable Factual 21 Jul 2017 Amodei said he thinks OpenAI has generally been very selective in whom it picks. Factual · 21 Jul 2017 Their own work or organisation
Unresolvable Prediction 17 Aug 2015 Amodei guessed that most people would support embryo screening for severe or debilitating mental illness. Prediction · 17 Aug 2015 Philosophy, ethics and society
Unresolvable Prediction 17 Aug 2015 Amodei said he suspects AI-accelerated biology will greatly expand what is possible, with weight, physical appearance, reproduction and other biological processes becoming fully under people's control. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 The rate of discovery of key biology and medicine technologies could be increased by 10x or more with many more talented, creative researchers. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 Dario Amodei suspects that some new and stranger economic arrangement will be needed, one that no one today has done a good job of envisioning. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Prediction 17 Aug 2015 Amodei said the cause of Alzheimer's seems like the type of problem that better measurement tools isolating biological effects could solve, and that he is bullish about AI's ability to solve it. Prediction · 17 Aug 2015 AI capabilities
Unresolvable Prediction 17 Aug 2015 Amodei suspects that improved mental health will ameliorate many other societal problems, including ones that seem political or economic. Prediction · 17 Aug 2015 Philosophy, ethics and society
Unresolvable Prediction 17 Aug 2015 Amodei said it seems very likely that there is room for smarter-than-human AI systems to do a better job than humans at the epidemiological modeling used in disease eradication campaigns. Prediction · 17 Aug 2015 AI capabilities
Unresolvable Factual 17 Aug 2015 Successful biology projects are often scrappy or afterthoughts not initially thought promising, rather than massively funded efforts. Factual · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 There is a good chance Alzheimer's can eventually be prevented with relatively simple interventions once what is going on is actually understood. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 AI could enable reductions of 95% or more in both cancer mortality and incidence. Prediction · 17 Aug 2015 AI capabilities
Unresolvable Prediction 17 Aug 2015 Amodei expects improvements in mental health, well-being and education to increase democracy, citing that all three are negatively correlated with support for authoritarian leaders. Prediction · 17 Aug 2015 Philosophy, ethics and society
Unresolvable Prediction 17 Aug 2015 The basic development of AI technology and many, though not all, of its benefits seems inevitable unless the risks derail everything, and is fundamentally driven by powerful market forces. Prediction · 17 Aug 2015 AI timelines and forecasting
Unresolvable Prediction 17 Aug 2015 Amodei strongly suspects that details of individual neuron communication will be abstracted away in most interesting questions about computation and circuits. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 Amodei said he does not think governments will (or should) turn over their economic policy to an AI, even if it could do so; he also said he is somewhat skeptical that an AI could solve the socialist calculation problem. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Factual 17 Aug 2015 In recent years GWAS studies seem to suggest that correlations between psychiatric disease genes and positive traits might have been overstated. Factual · 17 Aug 2015 Science and research methods
Unresolvable Factual 17 Aug 2015 Over the last 30 years, mathematicians, computer scientists and physicists applying their skills to biology have been quite successful but have not had the truly transformative impact initially hoped for. Factual · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 Greatly improved embryo screening will likely make it possible to prevent most genetic disease. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 In the short term, comparative advantage will keep humans relevant and increase their productivity as AI advances. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Prediction 17 Aug 2015 It is plausible that "AI finance ministers and central bankers" could replicate or exceed the "10% accomplishment" that the surrounding text credits to human economic planners whose policy choices helped countries catch up with the developed world. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Prediction 17 Aug 2015 Once AI is broadly effective and cheap, the current economic setup will no longer make sense and a broader societal conversation on organizing the economy will be needed. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Prediction 17 Aug 2015 AI-enabled biological science may reduce the need for iteration in clinical trials by developing more accurate animal and cell models or simulations. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 The triumph of liberal democracy and political stability is not guaranteed and perhaps not even likely. Prediction · 17 Aug 2015 Philosophy, ethics and society
Unresolvable Prediction 17 Aug 2015 Amodei said he is open to the perhaps absurd-sounding idea that 1000 years of progress could happen in 5-10 years, but very skeptical that 100 years of progress could happen in 1 year. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 Once human lifespan reaches 150, longevity 'escape velocity' may be reached so most people alive today could live as long as they want. Prediction · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 In the long run AI will become so broadly effective and cheap that human comparative advantage will no longer keep the current economy sensible. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Factual 17 Aug 2015 Markets function better in the developed world and are typically good at bringing down the cost of high-value technologies over time. Factual · 17 Aug 2015 Economics and markets
Unresolvable Factual 17 Aug 2015 There is perhaps about one major biology discovery per year, and collectively these arguably drive over 50% of progress in biology. Factual · 17 Aug 2015 Science and research methods
Unresolvable Prediction 17 Aug 2015 There is good reason to think AI-enhanced research will give us the means to make mitigating climate change far less costly and disruptive. Prediction · 17 Aug 2015 AI capabilities
Unresolvable Prediction 17 Aug 2015 AI can likely do disease eradication and economic development tasks better than they are currently done. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Prediction 17 Aug 2015 If 20% annual GDP growth were achieved in the developing world, sub-Saharan Africa would reach China's current per-capita GDP in 5-10 years. Prediction · 17 Aug 2015 Economics and markets
Unresolvable Prediction 17 Aug 2015 Humans are likely to keep a relative or even absolute advantage over AI in the physical world for a significant time. Prediction · 17 Aug 2015 AI, the economy and work
Unresolvable Factual 17 Aug 2015 Dario Amodei thinks and talks a lot about the risks of powerful AI. Factual · 17 Aug 2015 Their own work or organisation
Back to the list

Dario Amodei

Open its page

Prediction AI safety and alignment

“I think it's going to come first from misuse, the biorisk stuff that I talked about.”

Dario Amodei, Anthropic CEO 8 August 2023 Dario Amodei (Anthropic CEO) - $10 Billion Models, OpenAI, Scaling, & AGI in 2 years (Dwarkesh Podcast)

Serious danger from scaled AI models will first come from misuse, such as biorisk, rather than misalignment.

Unresolvable The public record could not settle it.

The second judge would not resolve this claim; the verdict shown is the first judge's.

What would count as true or false

Judge as of end of 2025. The claim is that the first serious danger from scaled AI models comes from misuse (for example biological or chemical weapons uplift) rather than misalignment. A 'serious danger' means either an actual serious harm or a frontier lab or government formally finding that a model crossed a dangerous-capability threshold that requires heightened safeguards. True: the first such dangers recognised by 2025 were misuse-related (for example safeguards activated over CBRN uplift, or documented serious misuse incidents), and no misalignment-caused serious danger came earlier. Mostly true: misuse dangers clearly came first, but misalignment concerns (for example deception found in evaluations) arose around the same time without causing serious harm or formal threshold findings. Mixed: serious misuse and misalignment dangers were recognised at roughly the same time, with no clear first. Mostly false: no serious danger of either kind had arisen by end of 2025, but the misuse concerns that did emerge were only minor or early. False: the first serious danger recognised by 2025 came from misalignment (autonomous harmful behaviour of a model) rather than misuse.

Evidence

  1. Activating AI Safety Level 3 protections Anthropic · 22 May 2025 Own record
    “And indeed, in the lead up to releasing Claude Opus 4, we proactively decided to launch it under the ASL-3 Standard.”
  2. Introducing ChatGPT agent: bridging research and action OpenAI · 17 July 2025
    “While we don’t have definitive evidence that the model could meaningfully help a novice create severe biological harm—our threshold for High capability—we are exercising caution and implementing the needed safeguards now.”
  3. Agentic misalignment: LLMs as insider threats Anthropic · 20 June 2025 Own record
    “In at least some cases, models from all developers resorted to malicious insider behaviors when that was the only way to avoid replacement or achieve their goals—including blackmailing officials and leaking sensitive information to competitors.”
  4. Detecting and countering misuse of AI: August 2025 Anthropic · 27 August 2025 Own record
    “The actor targeted at least 17 distinct organizations, including in healthcare, the emergency services, and government and religious institutions.”

Reasoning

The only independent evidence (E2) states OpenAI lacked definitive evidence that its biological-harm threshold was crossed, so it does not show a formal threshold finding; E1 does not say what ASL-3 covers or that a threshold was found, and E3 and E4 do not establish the setting or severity of the reported behaviour or harm. The evidence therefore does not establish that misuse dangers clearly came first under the criteria.

Cite this claim

Dario Amodei, “Serious danger from scaled AI models will first come from misuse, such as biorisk, rather than misalignment.” (said 8 August 2023). Verdict: unresolvable. Reliability API, claim c_eb261a2b1cdc, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_eb261a2b1cdc

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.