Citation

Max Tegmark - Footnote 27

partial85% confidence

1 evidence check

Last checked: 4/3/2026

The claim mentions specific model versions (Claude 4 Opus, Gemini 2.5 Pro, o3, Grok 3) which are not explicitly mentioned in the source. The source only refers to the companies and their general AI models. The claim states the index was launched in 'summer 2025', but the source mentions both 'Summer 2025' and 'December 2025' editions.

Evidence — 1 source, 1 check

futureoflife.org/ai-safety-index-summer-2025/FLI(1 check)

partial85%Haiku 4.5 · 4/3/2026

Found: **AI Safety Index**: Launched in summer 2025, this initiative evaluates AI companies on their safety practices for models like <EntityLink id="E22" name="anthropic">Anthropic</EntityLink> Claude 4 Opu…

Note: The claim mentions specific model versions (Claude 4 Opus, Gemini 2.5 Pro, o3, Grok 3) which are not explicitly mentioned in the source. The source only refers to the companies and their general AI models. The claim states the index was launched in 'summer 2025', but the source mentions both 'Summer 2025' and 'December 2025' editions.

Debug info

Record type: citation

Record ID: page:max-tegmark:fn27