Apollo Research Scales Monitoring Control, Hires for Red Teaming and Safety
MariusHobbhahn · x · 2026-09-02
Apollo Research is doubling down on control and monitoring research, having recently successfully red-teamed Anthropic's auto-mode with plans for future campaigns across multiple labs.
Key Updates:
- Will soon start fine-tuning monitors at scale with strong preliminary results, which could be significant for controlling scheming behavior.
- Identified many low-hanging fruits in monitoring, with the main blocker being manpower.
Open Roles:
- Hiring for Research Scientist (Control), AI Red Team Engineer, and AI Security & Control Researcher.
- Positions open in London and San Francisco with visa sponsorship available.
More from Companies & People
- Hacktoberfest 2026 Shifts From Open-Source Contributions to Open AI Community Events — jonmarkgo · 2026-09-02
- Broadcom's VMware Private AI Push Signals Enterprise AI Moving From Pilots to Production — DavidLinthicum · 2026-09-02
- UT PGE Faculty Discuss AI Strategy for Teaching and Research — GeostatsGuy · 2026-09-02
- Meta's Return to AI Front Rank: Strategy and Stats — rohanpaul_ai · 2026-09-02
- Berkeley professor joins TransluceAI as senior research fellow — 2plus2make5 · 2026-09-02
- Pentium chip father Vinod Dham discusses AI and longevity — Chris_Armstrong · 2026-09-02