What I've learned doing this work, written for people who ship.
After a year of studying AI incidents, testing AI systems, and evaluating model outputs daily, I've noticed patterns. Most failures aren't sophisticated attacks. They're preventable. Someone just didn't ask the right questions.
Read Article →27 questions compiled from studying AI incidents and observing what separates successful deployments from failures. Not from consulting engagements. from research. From reading what went wrong and working backward to what should have been asked.
Read Article →I evaluate AI outputs from frontier models daily. Here's what actually works, what doesn't, and what the hype misses. The truth is more nuanced than either the AI doom or the AI utopia narratives.
Read Article →