X users clash over METR donor transparency as org's eval work resurfaces
austinc3301 · x · 2026-09-17
In an X exchange, @austinc3301 responds to @AndrewAzlan's criticism of AI safety eval org METR, noting the org publicly lists its donors and that US 501(c)(3) nonprofits aren't legally required to disclose every donation since many donors choose anonymity.
The linked METR about page highlights its core work: frontier model capability/risk evaluations, including the finding that AI agent task horizons double roughly every 7 months, a GPT-5.1-Codex-Max risk assessment, and an RCT showing experienced open-source developers systematically overestimate AI's productivity gains. Its Responsible Scaling Policies prototype has been adopted by nine AI developers.
Related event: METR's Funding Surges to $71M in Six Months, Sparking Transparency Debate(4 posts)→
More from AGI Musings
- So8res Adds One-Year Retrospective to 'If Anyone Builds It, Everyone Dies' E-book — connoraxiotes · 2026-09-17
- A Year After 'If Anyone Builds It, Everyone Dies', Authors Give Away 1,000 Free E-books — connoraxiotes · 2026-09-17
- 'This will kill everyone, give us more power': blogger mocks recurring AI doom rhetoric from the world's most powerful — kevinnbass · 2026-09-17
- Podcast: TheZhi and Robert Wright debate whether the AI 'slowdown' is real — TheZvi · 2026-09-17
- Ex-HRT quant: LLM agents now let average CS grads do elite quant research — igarciacamargo · 2026-09-17
- Marx's failed predictions mirror today's AI forecasts, argues one observer — kevinnbass · 2026-09-17