Thao (Amelia) Pham

Mathematics, Computer Science @ Berea College

I'm a final-year undergraduate student at Berea College, United States, majoring in Mathematics and Computer Science. My research concerns emergent deceptive misalignment in multi-agent interactions, oversight mechanisms to prevent loss of control, and evaluating cooperative properties of AI agents. In my past life, I was a Research Assistant at Vanderbilt University, supervised by Dr. Douglas Fisher, completed two Open Source Software Internships, and led an NGO for climate change.

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 thaoameliapham@gmail.com

Update

Aug 15, 2025

I joined Jinesis AI Lab as a part-time Pre-Doctoral Student Researcher, supervised by Zhijing Jin, working on Open Source Game Theory.

Jul 23, 2025

I gave a lightning talk on my on-going research on multi-agent deception at the Human-aligned AI Summer School 2025.

Jul 10, 2025

I presented my poster on my ongoing research on multi-agent deceptive behavior modeling at the Cooperative AI Summer School 2025.

May 30, 2025

I began my Open Philanthropy-funded research project on capability in multi-agent systems, mentored by Lewis Hammond (Cooperative AI Foundation).

Apr 4, 2025

Our paper, APriCoT, accepted at CogSci 2025.

Sep 20, 2024

Two papers, Base-Rate Effect on LLM Benchmark Performance and Large Language Model Recall Uncertainty, accepted to EMNLP 2024 and CoNLL 2024.

May 6, 2024

I began my research internship at Vanderbilt University in studying human-like LLMs, supervised by Dr. Douglas Fisher, Dr. Jesse Roberts, and Dr.Kyle Moore.