Neev Parikh

• Program

  Science of Trustworthy AI

• Institution

  METR

• Location

  USA

Neev is interested in understanding the capabilities, behavior and limitations of frontier models. At METR, Neev leads the deceptive capabilities workstream, which is focused on understanding if, and how well, models can take subversive actions in various settings.

He holds a MSc. in computer science from Brown University, where he conducted research on reinforcement learning. Prior to METR, he worked at Stripe on data and ML products.