AI safety researcher

Tomek Korbak

Google Scholar dblp

Last 12 months7 papers

Evaluation and benchmarks 2Defenses 2

AI safety papers from the last year

On this site: 15 papers

September 2026

Alignment Forecasting: Predicting Misalignment From Training Data
Sep 19Evaluation and benchmarksarXiv ↗
Inoculation Midtraining with Learned Neologisms
Sep 14DefensesarXiv ↗

March 2026

Reasoning Models Struggle to Control their Chains of Thought
Mar 5Evaluation and benchmarksarXiv ↗

February 2026

Training Agents to Self-Report Misbehavior
Feb 25DefensesarXiv ↗

December 2025

Authorship follows dblp and the author's own Google Scholar profile; only AI safety papers on this site are listed.