<p>In experiments in which physicians made decisions about treating hypothetical patients, the physicians tended to trust incorrect advice presented as being generated by artificial intelligence (AI), even after being given the opportunity to notice that patient recovery data contradicted the recommend</p>