
AI Frontiers with James Zou: The Future of Multi-Modal AI in Medicine
NEJM AI Grand Rounds
00:00
Behavior Changes of Chat GPT-4
The chapter discusses the study on behavior changes of Chat GPT-4 over time, focusing on safety, biases, and response capabilities. It highlights variations in answering sensitive and non-sensitive questions, as well as the application of reinforcement learning from human feedback in training these models.
Transcript
Play full episode