Ranking the week's takes on METR third-party evaluation, from sneers to COI claims
dgrobinson · x · 2026-09-18
- A ranking of this week's takes on METR's embedded third-party evaluation, from worst to best, earned a "10/10 no notes" endorsement.
- It mocks contentless sneers at the outgroup and conspiracy-style METR critiques backed only by a huge Sankey diagram.
- Higher-ranked takes trace specific chains of actors and funding relationships that would make strong conflicts of interest problematic.
More from Fun
- GPT-6 Astra autonomously cracks an unsolved 1941 Enigma message — npew · 2026-09-18
- Training a robotic hand for gestures produced a surprisingly heartwarming moment — techartist_ · 2026-09-18
- ChatGPT used to ask for clarification — now it just says goodbye — yuntiandeng · 2026-09-18
- More Space Cat video generation experiments with OpenArt — TheChuckTone · 2026-09-18
- "There's a 5% chance AI kills us all — but almost all 5% chances don't happen" — NathanpmYoung · 2026-09-18
- Satire: claiming AI will destroy humanity is just a relevance play for replaced experts — bindureddy · 2026-09-18