news

Feb 27, 2026 New preprint on “Pressure Reveals Character: Behavioural Alignment Evaluation at Depth” now available on arXiv!
Jan 29, 2026 Our paper “Predictable Artificial Intelligence” has been published in the Artificial Intelligence journal!
Jan 26, 2026 Our paper “Formalising Human-in-the-Loop: Computational Reductions, Failure Modes, and Legal-Moral Responsibility” has been accepted to ICLR 2026!
Jan 05, 2026 I have joined Prolific as an AI Research Engineer!
Dec 19, 2025 Our paper “General Scales Unlock AI Evaluation with Explanatory and Predictive Power” is in press at Nature.
Nov 05, 2025 Our paper on “Conversational complexity for assessing risk in large language models” has been published in EPJ Data Science!
Oct 22, 2025 New preprint on “I Spy With My Model’s Eye: Visual Search as a Behavioural Test for MLLMs” now available on arXiv!
Oct 10, 2025 Two reports on AI model categorisation frameworks for the EU AI Act have been published by the EU Publications Office!
May 16, 2025 Our survey on AI evaluation was accepted as IJCAI survey track paper!