Imperial College postdoc opening: detecting unverbalized dangerous capabilities in frontier models
oanacamb · x · 2026-10-24
A researcher at Imperial College London, funded by a personal grant, is hiring a postdoc in AI Safety & Alignment to work on "Detecting Unverbalized Dangerous Capabilities in Frontier Models."
- Application deadline: 22 November; start date ASAP
- Requirements: PhD (or near submission) with a strong publishing record in AI safety, alignment, or frontier model evaluations
More from Companies & People
- NVIDIA Isaac slide reveals full-stack robotics platform with 20+ partners — ___Mufasaa · 2026-10-25
- FBI Probes Ex-OpenAI Engineer for Allegedly Stealing Secrets for Saudi Arabia — Polymarket · 2026-10-25
- Anthropic shares a tag-able list of product leads for direct user feedback — alexalbert__ · 2026-10-25
- AI influencer reveals labs offered paid posts if he hid sponsorship tags — prasenx · 2026-10-25
- Vercel CEO rauchg on agents: 'It can explore the world on your behalf' — soleio · 2026-10-25
- France pushes out entrepreneurs while the US attracts them — bad news for Mistral — debreuil · 2026-10-25