Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
OpenAI safety Research leader who helped form ChatGPT Responses to users who are struggling Mental health crises She announced her departure from the company internally last month, WIRED has learned. Andrea Vallone, head of the safety research team known as Model Policy, is set to leave OpenAI at the end of the year.
OpenAI spokeswoman Kayla Wood confirmed Vallone’s departure. Wood said OpenAI is actively looking for a replacement, and in the meantime, Vallone’s team will report directly to Johannes Heidke, the company’s head of safety systems.
Vallone’s departure comes as OpenAI faces increasing scrutiny over how its flagship product responds Users are in distress. In recent months, several lawsuits have been filed against OpenAI claiming that users formed unhealthy attachments to ChatGPT. Some lawsuits allege that ChatGPT contributed to mental health breakdowns or encouraged suicidal thoughts.
Amid this pressure, OpenAI is working to understand how ChatGPT handles annoyed users and improve chatbot responses. Model Policy is one of the teams leading this work, and leads October report Details the company’s progress and consultations with more than 170 mental health experts.
In the report, OpenAI said hundreds of thousands of ChatGPT users could emerge Signs of a manic or psychotic crisis Every week, more than one million people “have conversations that include clear indications of potential suicidal planning or intent.” By updating GPT-5, OpenAI said in the report that it was able to reduce spam responses in these conversations by 65 to 80 percent.
“Over the past year, I have led OpenAI research on an almost unprecedented question: How should models respond when faced with signs of emotional overdependence or early indicators of a mental health disorder?” Fallone wrote in A mail He is linkedin.
Vallone did not respond to WIRED’s request for comment.
Making ChatGPT fun to chat with, but not overly so, is one of the core tensions of OpenAI. The company is aggressively trying to expand ChatGPT’s user base, which now includes more than 800 million people weekly, to compete with AI-powered chatbots from Google, Anthropic, and Meta.
After OpenAI launched GPT-5 in August, users pushed back, arguing that the new model was too late Surprisingly cool. In the latest update to ChatGPT, the company said it has significantly reduced adulation while maintaining the “warmth” of the chatbot.
Fallone’s exit follows August reorganized for another group Focus on ChatGPT responses to afflicted users, and typical behavior. Its previous leader, Joan Zhang, left that role to start a new team exploring new approaches to human-AI interaction. The remaining Model Behavior crew was transferred under the command of Max Schwarzer after training.