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?
79 72 85 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
80 72 88 -
Promises
Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Insufficient record: 1 promise resolved, 5 needed for a score.
All types combined: 79 (likely 74 to 84, high confidence), from 102 claims: 64 factual claims, 38 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Gary Marcus's claims
52 claims
“Many would agree we have reached a point of diminishing returns for scaling; faith in scaling as a route to AGI has dissipated.”
Gary Marcus, author January 2026 Six (or seven) predictions for AI 2026 from a Generative AI realist (Marcus on AI)
Many would agree the AI field has reached a point of diminishing returns from scaling, and faith in scaling as a route to AGI has dissipated.
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.
What would count as true or false
Judge as of January 2026. The claim is about degree and has two parts: (a) many people in the AI field agree that scaling (more compute, data and parameters for pretraining) has reached diminishing returns; (b) faith in scaling as a route to AGI has dissipated, meaning it has broadly faded among researchers, commentators and industry, not just among critics. TRUE: evidence from before or around January 2026 shows both. For (a), surveys of AI researchers or public statements from a substantial number of prominent researchers or lab leaders acknowledge diminishing returns from scaling. For (b), there is a clear, widely reported shift away from 'scaling alone gets us to AGI', with a majority or a clear plurality of experts in surveys doubting it, or leading proponents publicly stepping back. MOSTLY TRUE: (a) is well supported and faith in pure scaling has clearly declined, but sizeable parts of the field (for example several major labs) still openly state that scaling is a main path to AGI, so 'dissipated' slightly overstates the change. MIXED: one part holds and the other does not, or the evidence shows the field roughly split on both. MOSTLY FALSE: only a minority of critics hold these views, and the leading labs and most surveyed experts still publicly expect scaling to deliver large gains and to lead to AGI; the claim inflates a minority view into a broad consensus. FALSE: evidence from the time shows broad agreement that scaling returns were not diminishing and that confidence in scaling toward AGI was steady or rising.
Evidence
- AI scientists are sceptical that modern models will lead to AGI New Scientist · 14 March 2025
“In a survey of 475 AI researchers, about 76 per cent of respondents said it was “unlikely” or “very unlikely” that scaling up current approaches will succeed in achieving AGI.”
- AI Beyond the Scaling Laws HEC Paris (DARE) · 1 December 2025
“The scaling laws that powered the exponential progress of Large Language Models (LLMs) like GPT-4, and fueled bold predictions of Artificial General Intelligence (AGI) by 2026 from Sam Altman (OpenAI) and Dario Amodei (Anthropic) have started to show diminishing returns.”
- Ilya Sutskever — We're moving from the age of scaling to the age of research Dwarkesh Podcast · 25 November 2025
“But is the belief that if you just 100x the scale, everything would be transformed? I don’t think that’s true. So it’s back to the age of research again, just with big computers.”
- Scaling Could Now Give Diminishing Returns, We're Back To The Age Of Research: Ilya Sutskever OfficeChai · 25 November 2025
“Interestingly, Sutskever’s comments come at at time when Google researchers have said that they’re still seeing gains from pre-training.”
Reasoning
A survey of 475 AI researchers found about 76% judged it unlikely that scaling up current approaches will achieve AGI (E1), and an HEC Paris analysis said scaling laws had started to show diminishing returns (E2). Ilya Sutskever publicly said that 100x scale would not transform everything and that it is "back to the age of research" (E3). However, Google researchers were reported to be still seeing gains from pre-training (E4). "Dissipated" therefore slightly overstates how far the change went.
Cite this claim
Gary Marcus, “Many would agree the AI field has reached a point of diminishing returns from scaling, and faith in scaling as a route to AGI has dissipated.” (said January 2026). Verdict: mostly true. Reliability API, claim c_35e0bd10e1ad, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_35e0bd10e1ad
Claims without a verdict
Is a verdict here wrong, or do you speak for Gary Marcus? 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.