Articles

AI Passed the Exam. What Did It Understand?

In 2025, an AI model built by researchers at the University of Buffalo outperformed most human physicians on the United States Medical Licensing Examination. OpenAI’s o1 model scored 96% on MedQA, a benchmark drawn from the same exam. DeepSeek hit 93% on the clinical knowledge portion. These aren’t outliers — across law, engineering, accounting, and philosophy, AI systems now match or exceed human performance on professional examinations.

What You're Really Afraid Of (It's Not AI)

A 2026 Gallup survey found that workers at companies that have already adopted AI are more anxious about losing their jobs than those at companies that haven't. The people most unsettled are not the ones who've been laid off. They're the ones still employed — watching the tools improve, still doing the work, still getting paid — and feeling something shift that a pay cheque doesn't address.

Who Is the Creator?

A few months ago, an AI-generated image won first place at a state art competition. The artist — if that's the right word — had typed a prompt into a machine, refined it over several hundred iterations, and submitted the result. The judges didn't know. When the truth came out, the reaction was swift and split down the middle. Half the internet said it was cheating. The other half said it was the future.