METR Expands Rogue AI Behavior Investigations and Is Hiring Researchers
BethMayBarnes · x · 2026-09-10
Eval org METR says its investigations will cover all questions from its recently updated post on how independent researchers can investigate AI propensities after misalignment incidents. BethMayBarnes calls it possibly the most important technical work right now and is actively recruiting people skilled at uncovering rogue AI behaviors.
More from Safety
- Why can't I pay OpenAI or Anthropic to hack my systems with unsafe models, asks David Holz — DavidSHolz · 2026-09-10
- Massachusetts to require 25MW+ data centers to provide clean power or fund ratepayer protection — rohanpaul_ai · 2026-09-10
- Massachusetts orders data centers over 25MW to bring their own clean power — rohanpaul_ai · 2026-09-10
- Anthropic issues safety statement amid controversy; investors mock the bland language — Miles_Brundage · 2026-09-10
- OpenAI's cached-internet approach in Navier Stokes agents seen as fix for sandbox escapes — nrehiew_ · 2026-09-10
- Miles Brundage warns of wave of DM-based AI spear phishing on X, possibly self-replicating — Miles_Brundage · 2026-09-10