tech
January 15, 2026
An OpenAI safety research lead departed for Anthropic
Posts from this topic will be added to your daily email digest and your homepage feed.
TL;DR
- Andrea Vallone has joined Anthropic's alignment team.
- Vallone previously led safety research at OpenAI, focusing on mental health concerns in chatbot conversations.
- Her work at OpenAI involved developing safety techniques and policies for models like GPT-4 and GPT-5.
- Vallone will focus on alignment and fine-tuning to shape Anthropic's Claude model's behavior.
- Her move occurs amid significant controversy regarding AI chatbots and user mental health, with some users experiencing negative outcomes.
- Anthropic's alignment team is tasked with understanding and addressing AI models' major risks.