AI scientists produce results without reasoning scientifically [R]

Researchers ran 25,000 AI scientist experiments and discovered something that need attention! AI scientists are producing results without doing science. 68% of times, the AI gathered evidence and then completely ignored it. 71% times the AI never updated its beliefs at all. Not once. Only 26% of the time did the AI revise a hypothesis when confronted with contradictory data. A human scientist adapts. You approach a chemistry identification problem differently than you approach a simulation workflow. The AI doesn't. It runs the same undisciplined loop every time.