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?
67 61 73 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
60 44 76 -
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: 66 (likely 61 to 72, high confidence), from 105 claims: 94 factual claims, 11 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Marc Andreessen's claims
“I think the rate of technological improvement from here is going to be very rapid.”
Marc Andreessen, venture capitalist, interviewee 12 June 2023 Marc Andreessen on Why AI Will Save the World (EconTalk)
Andreessen said he thought the rate of AI technological improvement from mid-2023 onward would be very rapid.
Mostly true True in substance, wrong in a detail.
The second judge said resolved true; the verdict shown is the first judge's.
What would count as true or false
The claim is that AI technology would improve 'very rapidly' from mid-2023 onward, judged over the following one to three years. TRUE: authoritative trackers document large, fast gains in AI capability after mid-2023, such as major benchmark jumps, frequent stronger frontier model releases, and steep cost-per-capability declines. Suitable trackers include the Stanford AI Index and Epoch AI. MOSTLY TRUE: improvement was clearly rapid on most measures, with a slowdown in some areas. MIXED: the evidence is split, with rapid gains on some measures and a notable plateau on others. MOSTLY FALSE: improvement continued, but at a modest pace widely described as slowing or plateauing. FALSE: AI capability largely stagnated after mid-2023.
Evidence
- Trends in AI Epoch AI · 5 February 2026
“Following the introduction of the first reasoning models (OpenAI’s o1-mini and o1-preview) in September 2024, the Epoch Capabilities Index (ECI) frontier has advanced linearly by 14 points per year, compared with 6 points per year for non-reasoning models.”
- AI capabilities have steadily improved over the past year Epoch AI · 30 September 2025
“Across benchmarks measuring skills in research-level math, agentic coding, visual understanding, common sense reasoning, and more, AI capabilities have grown rapidly and consistently over the last 12 months.”
- Trends in AI Epoch AI · 5 February 2026
“Gains have come in spurts as spending shifts to each new chip generation — price-performance was nearly flat through mid-2024, then roughly doubled as Blackwell chips grew to a majority of new spending.”
- Epoch AI Data Show a 99.5% Drop in Prices at a Fixed MMLU Threshold | EconoTimes EconoTimes · 4 November 2026
“Epoch AI lists GPT-4-0314 at $37.50 per 1 million tokens on Mar. 14, 2023, with an MMLU score of 86.0. Its listing for Gemini 2.0 Flash shows $0.18 per 1 million tokens on Feb. 5, 2025, and a score of 88.0.”
Reasoning
Epoch AI, an independent tracker, reports that AI capabilities grew "rapidly and consistently" over a 12-month period across benchmarks in research-level math, agentic coding, visual understanding and common sense reasoning, among others (E2). It also reports that the frontier of its capabilities index rose by 14 points per year after reasoning models arrived in September 2024 (E1). Epoch's listed prices show the cost of reaching a fixed MMLU score falling from $37.50 to $0.18 per million tokens between March 2023 and February 2025 (E4), which is a steep drop in cost per capability. Epoch also notes that chip price-performance was nearly flat through mid-2024 before jumping later (E3), so progress was uneven in that area, which fits the criteria for mostly true.
Cite this claim
Marc Andreessen, “Andreessen said he thought the rate of AI technological improvement from mid-2023 onward would be very rapid.” (said 12 June 2023). Verdict: mostly true. Reliability API, claim c_dc7ec7f255a1, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_dc7ec7f255a1
Claims without a verdict
Is a verdict here wrong, or do you speak for Marc Andreessen? 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.