Project Telos

AI & ML interests

Towards a Science of AI Agency: Modelling, Measuring, and Intervening on Goal-Directed Behaviour

Recent Activity

kortukov  updated a model 3 days ago

project-telos/interp

kortukov  authored a paper about 2 months ago

Strategic Dishonesty Can Undermine AI Safety Evaluations of Frontier LLM

gsarti  authored a paper 8 months ago

QE4PE: Word-level Quality Estimation for Human Post-Editing

View all activity

Team members 4