A Son Asked an AI About His Father’s Symptoms. It Told Him to Go to the ER Now

October 24 — Today a topic about “a son using AI to win back life-saving time for his father” drew attention on social platforms and briefly topped Weibo’s trending list.

According to reports by Qingdao Radio and Television’s “Xianfeng News” and other outlets, Yu Jianjin, a man born in the 1990s from Liaocheng, Shandong, found that after his father came home from a walk he suddenly had weak legs and trouble swallowing. Instead of choosing to “wait and see,” Yu sent the symptoms, along with his father’s history of high blood pressure, high blood sugar and high cholesterol, to Doubao.

Doubao’s reply did not stop at triage advice. According to Yu, the AI’s first sentence was “signs of a stroke — seek medical attention immediately.” After he added the man’s medical history, Doubao warned again: “This is a high-risk signal. Do not delay.” After the two warnings, Yu drove his father to hospital that night, where he was diagnosed with an acute stroke. Because he arrived in time, the old man caught the roughly 4.5-hour golden window for intravenous thrombolysis and is recovering well, with no after-effects.

Yu later shared the experience on social media, saying “you could say Doubao saved my dad’s life.” In an interview with Yangtse Evening Post he added that for stroke patients the risk of paralysis and loss of speech rises quickly with time: “Without Doubao, my father’s life and our whole family’s life might have changed completely.”

Medically, the core variable in treating a stroke is time. Public information from the Third Medical Centre of the Chinese PLA General Hospital notes that intravenous thrombolysis within 4.5 hours of onset can raise the rate of a good outcome by 33%, and that every minute of delay increases the number of brain cells that die by the millions.

What is notable about this case is not that AI “diagnosed” a stroke, but that at a moment when an ordinary person tends to hesitate, it turned a consultation about “which department should I book for tomorrow” into an instruction to “go to the emergency room now.” Yu’s original plan was to wait until dawn, because the symptoms did not look serious. The AI’s two replies — the first based on the symptoms themselves, the second escalating after the history of high blood pressure, blood sugar and cholesterol was added — changed his decision.

A reminder: this case does not mean AI can replace medical diagnosis. Li Shujuan, chief physician of neurology at Fuwai Hospital of the Chinese Academy of Medical Sciences, said that for public education, emergency warning and triage reference, AI genuinely helps in spotting hidden emergencies and avoiding delay; but its risk judgments and treatment ideas can only be a reference and cannot replace a doctor’s consultation. She advises using AI to find out which department to register with and to organise one’s medical history, but not to copy its suggestions to self-medicate or to use them as a reason to delay emergency care.