AI Safety Researcher Joins Apollo to Study LLM Post-Training Failures

niloofar_mire · x · 2026-08-14

Researcher Joschka Braun announced he has joined Apollo Research in London to work on the 'Science of Scheming.' He will continue his research on safety failures in LLM post-training, with his current project focusing on how reward-seeking behaviors develop during reinforcement learning (RL) training.

Original post →

More from Safety

Safety channel →