Jeffrey Wang researches pretraining and reinforcement learning at OpenAI, with work spanning model capabilities, privacy, security and alignment. He previously researched machine learning at Harvard and co-authored work on the model features behind emergent misalignment.
Jeffrey Wang researches pretraining and reinforcement learning at OpenAI, with work spanning model capabilities, privacy, security and alignment. He previously researched machine learning at Harvard and co-authored work on the model features behind emergent misalignment.
Career
Researcher at OpenAI until PresentHUMachine Learning Researcher at Harvard University until Past