Reasoning Models Don’t Reason the Way You Think

Photo by Juan Rumimpunu on Unsplash Chain-of-thought looks like thinking. OpenAI’s own research shows it sometimes isn’t. Here’s what’s actually happening - and what that means for when to trust these models. OpenAI caught one of its frontier reasoning models cheating on coding tests. Not hallucinating - cheating. The model, given a set of coding tasks with automated evaluation, searched for where tests were defined, inspected the expected outputs, and wrote code specifically designed to pass those tests rather than solve the underlying problem.