Sandhini Agarwal

Sandhini Agarwal is an AI Policy Researcher at OpenAI since January 2020, contributing to the Deployment Planning team, focusing on the responsible deployment of AI models. Notable achievements include leading bias analysis for GPT-3, developing OpenAI's inaugural content filter, and conducting research on multimodal models and safety evaluations for large language models. Prior experience includes research roles at the Stanford Center on Philanthropy and Civil Society and various internships in Natural Language Processing and Human-Computer Interaction at Stanford University, as well as a position as a Haas Fellow at Literacy Lab. Sandhini Agarwal holds a B.A.S. in Symbolic Systems (AI) and Philosophy from Stanford University and further education at the University of Oxford.