Thao Pham

Computational Math & Information Science @ Berea College

I'm a final-year undergraduate student at Berea College, United States. I'm concerned about risks from advanced AI systems, which I'm primarily motivated to study safety and cooperation in multi-agent settings. My research interests ponder questions like:

• How can we evaluate and detect multi-agent scheming ability?
  • Can we effectively model Bayesian Theory of Mind in multi-LLM-agents settings to study scheming behavior?
• How can we study LLM-based agents' cooperative properties, or steer their aspirations towards human values, in open-ended environments?

Besides research, I contribute to expand the AI Safety community, e.g., founding the AI Safety Initiative at Berea College, and facilitating for Center for AI Safety. For more details, please refer to my CV.

Feel free to contact me via phamt2@berea.edu, schedule a chat (I'm always looking for new research collaborators), or consider leaving me anonymous feedback.