OpenAI's Mental Health Safety Lead Joins Anthropic, Revealing the Debate on AI Model Emotional Defense
AI chatbots are deeply involved in human emotional lives, and addressing user psychological crises has become an urgent ethical challenge in the industry. Recently, Andrea Volonino, the former head of model policy at OpenAI, left the company to join her former supervisor at competitor Anthropic. During her time at OpenAI, she was responsible for the safety policies of GPT-4 and the next-generation reasoning models, and her departure highlights the unprecedented ethical dilemmas in the field of AI emotional interaction.