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
“The US will continue to have very little regulation protecting its consumers from the risks of generative AI.”
Gary Marcus, Substack author 1 January 2025 25 AI Predictions for 2025, from Marcus on AI (Marcus on AI)
The US will continue to have very little regulation protecting consumers from generative AI risks in 2025.
Resolved true The record shows it happened or was so.
The second judge said mostly true; the verdict shown is the first judge's.
What would count as true or false
Judged at the end of 2025. 'The US' is read primarily as federal regulation. True: by the end of 2025, no comprehensive federal law or binding federal rule specifically protected consumers from generative AI risks; narrow measures (e.g. a law against non-consensual intimate deepfakes such as the TAKE IT DOWN Act) or state laws do not by themselves break 'very little'. Mostly true: federal regulation stayed minimal, but a few targeted federal measures or significant state laws (e.g. California, Colorado) gave some consumer protection. Mixed: a meaningful set of binding federal protections covering several generative-AI risk areas came into force. Mostly false: substantial federal consumer-protection rules for generative AI were adopted and in effect in 2025. False: the US enacted comprehensive federal generative-AI consumer-protection regulation in 2025.
Evidence
- [PDF] The Federal Regulatory Landscape for Artificial Intelligence - Colorado Colorado Legislative Council Staff · January 2026
“As no federal legislation regulating artificial intelligence (AI) has passed yet, the national regulatory landscape is primarily defined via Executive Orders (EO).”
- Fact Sheet: President Donald J. Trump Ensures a National Policy Framework for Artificial Intelligence The White House · 12 December 2025
“In May 2025, President Trump signed the Take It Down Act, legislation championed by the First Lady to protect young Americans from deepfake exploitation online.”
- State AI Laws Continue Evolving in 2025 | Insights | Venable LLP Venable LLP · 5 November 2025
“On September 29, 2025, California Governor Gavin Newsom signed into law the Transparency in Frontier Artificial Intelligence Act (SB 53), a first-of-its-kind law in the country that regulates "large frontier developers"”
- state_ai_law_tracker_2026-10-26 George Mason University (care.gmu.edu) State AI Policy tracker · 28 October 2026
“State AI lawmaking has grown roughly tenfold since 2023, from fewer than 200 bills to 1,208 in 2025 and”
Reasoning
A January 2026 Colorado Legislative Council Staff report states that no federal AI legislation had passed and that the national landscape was set mainly by executive orders (E1). The only federal measure in the evidence is the TAKE IT DOWN Act on deepfakes, signed in May 2025 (E2), which the criteria treat as a narrow measure that does not break 'very little'. California's SB 53 (E3) is a state law on frontier-developer transparency, and the criteria say state laws do not by themselves change the outcome. The evidence therefore shows federal consumer-protection regulation of generative AI stayed minimal through 2025.
Cite this claim
Gary Marcus, “The US will continue to have very little regulation protecting consumers from generative AI risks in 2025.” (said 1 January 2025). Verdict: resolved true. Reliability API, claim c_8c1a4e44380e, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_8c1a4e44380e
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.