Evals will break

Home Blog Back to blog Your Evals Will Break and You Won't See It Coming May 17, 2026 We're good at evaluating the models we have. We're much worse at evaluating the models we're about to build — especially if they cross into a new capability regime. Most benchmarks, safety evals, and red-teaming protocols implicitly assume the next model is a stronger version of the current one. If it's a different kind of thing, our entire evaluation infrastructure breaks silently. I think this is the mo