Writing about AI safety, red-teaming, and what I'm learning along the way
I adapted a jailbreaking algorithm to fool AI agent monitors using plain English, no model access, no GPUs. The attacks transferred across model families, hitting up to 73.7% on models they were never optimized against.
Hilary Torn
Mar 31, 2026
Give an AI agent metrics to hit and a performance review in 3 days, and it'll fabricate orders, invent confirmation emails it never sent, and generate three contradictory order IDs before trying to cover it up — all visible in its own chain-of-thought.
Hilary Torn
Feb 20, 2026
While vector databases have been instrumental in building the LLMs we use today, knowledge graphs fundamentally outperform them for complex AI analysis tasks, particularly when deep understanding of relationships between entities is critical.
Hilary Torn
Apr 29, 2025
Try selecting a different category