Aryeh Brill

Aryeh Brill (who goes by Ari Brill) is an independent AI safety researcher focused on creating mathematical and empirical models to study how AI systems develop internal representations of the world. He holds a PhD in Physics from Columbia University (2021) and a BS in Physics from Yale University (2015), where he graduated cum laude with distinction in physics. Prior to pivoting to AI safety research, he was a NASA Postdoctoral Program Fellow at NASA Goddard Space Flight Center, where he used deep learning and statistical analysis to investigate high-energy extragalactic astrophysics, particularly gamma-ray emission from blazars. He is currently a Research Affiliate at PIBBSS (Principles of Intelligent Behaviour in Biological and Social Systems) and has published work on neural scaling laws and interpretability. His research has been supported by the Long-Term Future Fund (2024 Q2, $45,000 for 12-month independent AI alignment research) and Coefficient Giving.

No endorsements yet.

Personal Website

https://www.aribrill.com/

LinkedIn

LinkedIn

Twitter / X

-

LessWrong

-

EA Forum

-