Dario Amodei: checked claims

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.

See the factual claims
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.

See the predictions
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.

See the promises

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

17 claims

17 claims, sorted by newest said. Choose a claim to read it in full.
Verdict	Kind	Said	Claim	Topic
Mostly false	Prediction	23 Sep 2026	Amodei said that "we" believe AI for biology is on an exponential trend similar to the one he described for AI in mathematics. Prediction · 23 Sep 2026	AI capabilities
Mostly true	Prediction	28 Feb 2026	Anthropic will be fine, and the impact of the supply chain risk designation on it will be fairly small. Prediction · 28 Feb 2026	AI policy and governance
Resolved true	Prediction	10 Mar 2025	US export controls may have the effect of stimulating China's chip development. Prediction · 10 Mar 2025	AI policy and governance
Resolved false	Prediction	10 Mar 2025	Within three to six months of March 2025, AI will be writing 90 percent of code. Prediction · 10 Mar 2025	AI timelines and forecasting
Resolved false	Prediction	10 Mar 2025	Within twelve months of March 2025, AI may be writing essentially all of the code. Prediction · 10 Mar 2025	AI timelines and forecasting
Mostly true	Prediction	10 Mar 2025	Anthropic's models may reach AI Safety Level 3 in 2025. Prediction · 10 Mar 2025	AI timelines and forecasting
Mostly true	Prediction	28 Feb 2025	In the short run (about 6 to 12 months), AI will augment and increase coders' productivity rather than replace them. Prediction · 28 Feb 2025	AI, the economy and work
Resolved true	Prediction	28 Feb 2025	By the end of 2025, AI will do very serious things in coding. Prediction · 28 Feb 2025	AI capabilities
Mostly false	Prediction	11 Nov 2024	Amodei said that, after progress from 3% to 50% on SWE-bench in 10 months, state-of-the-art models would probably be at about 90% in another year, while adding that he did not know and it might be less. Prediction · 11 Nov 2024	AI capabilities
Mostly true	Prediction	11 Nov 2024	Amodei guessed that by about November 2025, AI models would be able to perform computer-use tasks very, very reliably. Prediction · 11 Nov 2024	AI capabilities
Resolved true	Prediction	11 Nov 2024	Dario Amodei said he would not be surprised at all if Anthropic's models hit ASL-3 in 2025. Prediction · 11 Nov 2024	AI safety and alignment
Resolved false	Prediction	11 Nov 2024	Amodei guessed that within about 10 months AI models would probably get pretty close on typical real-world programming tasks, reaching at least 90%. Prediction · 11 Nov 2024	AI capabilities
Resolved true	Prediction	8 Aug 2023	In a year or two (from 2023), AI model consciousness might become a very real concern. Prediction · 8 Aug 2023	Philosophy, ethics and society
Mostly true	Prediction	8 Aug 2023	AI models will not have enough autonomy within two years to raise alignment concerns, though that may not be far behind. Prediction · 8 Aug 2023	AI safety and alignment
Resolved false	Prediction	13 Oct 2022	The situation where several companies can produce good reasoning models will rapidly cease as everyone moves further up the scaling curve. Prediction · 13 Oct 2022	Economics and markets
Mostly true	Prediction	17 Aug 2015	AI seems likely to enable much better propaganda and surveillance, which are major tools of autocrats. Prediction · 17 Aug 2015	AI policy and governance
Mostly true	Prediction	17 Aug 2015	Climate change will be felt much more strongly in the developing world, hampering its development. Prediction · 17 Aug 2015	Other global risks

Back to the list

Dario Amodei

Open its page

Prediction AI safety and alignment

“I would not be surprised at all if we hit ASL-3 next year.”

Dario Amodei, Anthropic CEO 11 November 2024 Transcript for Dario Amodei: Anthropic CEO on Claude, AGI & the Future of AI & Humanity (Lex Fridman Podcast)

Dario Amodei said he would not be surprised at all if Anthropic's models hit ASL-3 in 2025.

Resolved true The record shows it happened or was so.

The evidence is the person's or their organisation's own records.

Said for an organisation they lead or work for; it counts on both records.

What would count as true or false

The test is whether, between 1 Jan and 31 Dec 2025, Anthropic formally determined that one of its models required ASL-3 protections under its Responsible Scaling Policy. This includes a precautionary determination that ASL-3 capability could not be ruled out, followed by activation of the ASL-3 deployment and/or security standard. TRUE: such a determination or activation is dated within 2025. MOSTLY TRUE: the activation happened in 2025, but Anthropic described it as covering only one of its ASL-3 capability thresholds (for example CBRN only) rather than all of them. MIXED: not normally used, because this is a single-outcome prediction. MOSTLY FALSE: the first ASL-3 determination came only after 31 Dec 2025, or it came before 2025 (so the model did not 'hit' it next year). Also mostly false if the only 2025 events were internal statements that a model was 'close to' ASL-3, with no activation. FALSE: Anthropic records from 2025 or later (RSP capability reports, system cards) show that every model deployed in 2025 stayed under ASL-2 with no ASL-3 determination.

Evidence

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.”
ActivatingASL-3_2025 Anthropic · May 2025 Own record
“the ASL-3 Standard, and we are implementing it now as a precautionary measure.”
Claude 4.1 System Card Anthropic · August 2025 Own record
“Like Claude Opus 4, Claude Opus 4.1 is deployed under the AI Safety Level 3 (ASL-3)”
Anthropic’s Transparency Hub Anthropic Own record
“Based on our evaluations detailed below, we ruled out the need to release Claude Haiku 4.5 under the ASL-3 Standard and are instead releasing it under the ASL-2 Standard.”

Reasoning

Anthropic's official announcement dated 2025-05-22 [E1] says it proactively chose to launch Claude Opus 4 under the ASL-3 Standard. The accompanying May 2025 report [E2] says ASL-3 is being implemented now as a precautionary measure. Both match the criteria's precautionary-determination-plus-activation case within 2025, and the August 2025 Claude 4.1 System Card [E3] confirms ASL-3 deployment continued. These are Anthropic's own official records, and the quoted text does not say the activation was limited to a single capability threshold, so the evidence supports true rather than mostly true.

Cite this claim

Dario Amodei, “Dario Amodei said he would not be surprised at all if Anthropic's models hit ASL-3 in 2025.” (said 11 November 2024). Verdict: resolved true. Reliability API, claim c_4e404ff97d27, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_4e404ff97d27

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.