Blog

2026-06-22 Mechanistic Interpretability as a Security Tool

2026-06-20 Prompt Injection, 2022 vs Today: A Retrospective

2026-06-18 The Format-Reliability Gap: Diagnosing and Repairing Insecure Code Generation

2026-06-15 Does AI Make You Write Insecure Code? A User Study

2023-01-01 Adversarial Fineturning against Prompt Injection Attacks