Where does registration take place?

Registration or viewing is completed on UK AI Forum's website. UPCOMING sends visitors using referral parameters.

Loading page…

Webinar

Anti-Scheming: how would we train a model not to scheme? (Bronson Schoen and Marius Hobbhahn, Apollo Research) — UK AI Forum

Name: Anti-Scheming: how would we train a model not to scheme? (Bronson Schoen and Marius Hobbhahn, Apollo Research) — UK AI Forum
Start: 2025-11-08
End: 2025-11-08

Saturday, November 8, 2025Webinar · Constance Tipper Lecture Theatre, Department of Engineering, Trumpington Street Cambridge, England, CB2 1PZ United Kingdom

About this event

Bronson Schoen (senior research engineer, Apollo Research) and Marius Hobbhahn (CEO, Apollo Research) will present Apollo’s & OpenAI’s joint work: ‘Stress Testing Deliberative Alignment for Anti-Scheming Training’. They attempted to use deliberative alignment to teach o3 and o4-mini

Anti-Scheming: how would we train a model not to scheme? (Bronson Schoen and Marius Hobbhahn, Apollo Research) — UK AI Forum

About this event

Topics covered

Frequently asked questions

About this event

Topics covered

Frequently asked questions

More from UK AI Forum

Loss-of-Control Governance: Degrees, Dynamics, and Preferences (Alejandro Ortega, Apollo Research) — UK AI Forum

AI & Societal Robustness Conference — UK AI Forum

AI Agents and Morality: Towards Norm-Aligned Design Choices (Esther Jaromitski, Queen Mary University) — UK AI Forum