Elon Musk has a new pitch, and it’s not rockets or robots this time. It’s about medicine. The man with every expensive physician at his disposal wants you to use Grok to handle your health issues instead.
In a resurfaced June 2025 clip that Musk amplified again on Jan. 1, he claimed Grok can analyze X-rays and MRI scans to deliver medical diagnoses. Sometimes better than doctors, he said. He doubled down by sharing a viral anecdote about a Norwegian man whose appendicitis was allegedly flagged by Grok. Musk also urged people to “Try it,” while implying that when doctors miss things, AI won’t.
His statement can be taken as optimism for artificial intelligence. But medical diagnosis concerns life and death. If we look at it that way, what Musk advertised comes off as a dangerously reckless framing. The Norway story that Musk keeps using as a basis for his claims was first shared on Reddit. While it does show something real if believed, it’s far less dramatic than Musk suggests.
A patient, still in pain after being discharged, described his symptoms to Grok. The AI suggested possible appendicitis or a perforated ulcer. The patient went back to the hospital. A CT scan performed by doctors confirmed appendicitis, and surgeons saved his life. In the story, what Grok did at best was prompt a second opinion. It cannot be seen as a replacement for a physical diagnosis.
Musk essentially wants to gamble with public health to test Grok
Medical experts have repeatedly warned that interactive AI like Grok and ChatGPT are not diagnostic tools. They don’t examine patients, can’t assess nuance, and don’t understand probabilities the way clinicians do. They are also prone to confident errors.
All other AI chatbot companies like OpenAI, Google, and Meta explicitly caution against using their models for medical diagnosis. But Musk went the opposite way. He even framed Grok’s abilities as sometimes “better than doctors.” The claim stands unsupported by evidence, peer-reviewed studies, regulatory approval, or clinical trials. It’s only marketing, and a dangerous one at that.
It’s important to understand that AI systems like Grok are not FDA-approved medical devices. More importantly, they are not legally accountable for harm. So if someone uses it for medical diagnosis, the consequences will fall on patients, not Silicon Valley founders.
Musk has a history of regularly positioning institutions as outdated gatekeepers who should be bypassed in favor of tech products. Medicine, apparently, is next on his target. But don’t be fooled. Even in Musk’s favorite example, a human doctor ultimately did their job. And these billionaires themselves go to Harvard doctors at the first sign of any inconvenience.
Have a tip we should know? [email protected]
