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?
83 78 88 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
Insufficient record: 2 predictions resolved, 5 needed for a score.
-
Promises
Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Insufficient record: 0 promises resolved, 5 needed for a score.
All types combined: 83 (likely 78 to 88, high confidence), from 86 claims: 86 factual claims. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Daniela Amodei's claims
38 claims
“And we’ve done this with a fraction of the compute our competitors have,” Amodei says.”
Daniela Amodei, Anthropic president, interviewed 27 January 2026 Anthropic cofounder Daniela Amodei says trusted enterprise AI will transcend the hype cycle (Fast Company)
Anthropic achieved its revenue growth using a fraction of the compute its competitors have.
Mostly true True in substance, wrong in a detail.
The second judge would not resolve this claim; the verdict shown is the first judge's.
Said for an organisation they lead or work for; it counts on both records.
What would count as true or false
True if credible evidence as of 2026-01-27 (company disclosures, reporting on compute capacity and spending, analyst estimates) shows that Anthropic achieved its revenue growth while having markedly less compute than its main competitors (e.g., OpenAI, Google), with its compute a clear minority share of theirs. Mostly true if Anthropic's compute was clearly smaller but 'a fraction' somewhat overstates the gap. Mixed if it had less compute than some competitors and comparable compute to others. Mostly false if the gap was small, or the claim rests only on comparison with the very largest competitor. False if Anthropic had compute comparable to or greater than its main competitors.
Evidence
- Anthropic is all in on 'AI safety'—and that's helping the $183 billion startup win over big business | Fortune Fortune · 2 December 2025
“Anthropic is a minor player, comparatively, in terms of our actual compute," she says.”
- Anthropic is all in on 'AI safety'—and that's helping the $183 billion startup win over big business | Fortune Fortune · 2 December 2025
“it’s only a third of the $235 billion OpenAI was budgeting over that time frame, according to information it had given its own investors.”
- Epoch Reveals OpenAI's Computing Power Grew 17x in Two Years | AlphaSignal AlphaSignal · 9 September 2026
“| Lab | Median H100e | 90% CI | |---|---|---| | OpenAI | 1,743,000 | 1.25M-2.19M | | Google DeepMind | 1,583,000 | 1.01M-2.55M | | Anthropic | 1,190,000 | 842K-1.72M |”
- ai-compute-users/docs/frontier-lab-compute-modeling-summary.md at main · epoch-research/ai-compute-users GitHub (epoch-research repository) · 13 July 2026
“Bottom line: ≥1M H100e end-2025 (conservative floor), likely <1.5M; ratio to OpenAI well over 50%, ~60% central.”
Reasoning
Epoch's modelling summary (E4) puts Anthropic's end-2025 compute at roughly 60% of OpenAI's, and at well over 50% even at the low end. That makes Anthropic clearly smaller, but its compute was not a clear minority share. Epoch's table (E3) also shows Anthropic below both OpenAI and Google DeepMind, though the quoted table does not state which period it covers. Fortune (E2) reports Anthropic's spending at about a third of OpenAI's budget, but that quote does not say clearly what "it" measures. Overall, Anthropic had markedly less compute than its competitors, but "a fraction" overstates the gap.
Cite this claim
Daniela Amodei, “Anthropic achieved its revenue growth using a fraction of the compute its competitors have.” (said 27 January 2026). Verdict: mostly true. Reliability API, claim c_01512983587c, method 0.1.0+28da144b. https://epistacklabs.com/claims/c_01512983587c
Claims without a verdict
Is a verdict here wrong, or do you speak for Daniela 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.