Ben Riley has spent the past few years warning strangers not to lean on artificial intelligence for life-or-death decisions—a job that now has poignant special meaning after his own father did exactly that and is no longer here to tell the tale. In a report by the New York Times, Teddy Rosenbluth traces how 75-year-old Joe Riley, a retired neuroscientist in Seattle, became convinced—largely with help from AI tools—that his doctors were wrong about his leukemia and that the cancer treatment they urged would only speed his decline. Joe obsessively queried Perplexity and other chatbots about his condition, generated a convincing-looking “research report,” and used it to overrule his oncologist, family, and even the scientists whose work the AI misquoted.

In Rosenbluth’s words: “‘Do you really think you know more than all of them because of this stupid AI report?’ Ben remembered asking. ‘Yes,’ Joe firmly responded.” By the time Joe finally agreed to treatment, he was too frail; he died in late 2025. Ben, who writes an AI-skeptic newsletter, doesn’t claim AI directly killed his dad, but he argues that flawed, authoritative-sounding outputs helped lock Joe into a fatal choice—despite his tech savviness and good medical care. “I will forever wonder whether my efforts came too late,” Ben wrote in a January essay. “There’s nothing I can do to change the past, of course. But I can for damn sure keep working to raise the consciousness of others.” The story lands as major tech companies, including Perplexity, roll out new AI health tools. Read the full account here for the details and ethical questions such use of AI raises.