Jiajia Li

When a soccer player heads the ball, the brain inside the skull jiggles slightly, like jelly in a sealed jar. A single impact leaves no trace, but hundreds repeat over the years. Doctors struggle to tell when these jolts start to cause harm: standard brain scans look normal, and the person may feel perfectly healthy. Scientists have found another way to peek into such a brain.

By 2025, large language models like GPT are no longer just chatting: they're helping scientists run experiments. But the same tool could fall into the wrong hands. A team of researchers built a digital proving ground to find out.