A woman from the U.S. claims that the chatbot developed by OpenAI, ChatGPT, saved her life by guiding her to the true cause of her symptoms: thyroid cancer that doctors had missed.
Lauren Bannon, a mother of two and marketing entrepreneur, divides her life between North Carolina and the U.S. Virgin Islands. In early 2024, she began experiencing morning stiffness in her fingers, which later escalated into severe stomach pains and sudden weight loss. Despite multiple doctor visits, she was diagnosed with rheumatoid arthritis—though her tests were negative—and her weight loss was attributed to simple acid reflux.
Frustrated by the lack of answers, Bannon decided to do what many in the digital age are beginning to consider an informal second opinion: she consulted ChatGPT, the language model developed by OpenAI. She had previously used it for work-related issues, but this time she employed it for a more personal search.
She asked what medical conditions might resemble rheumatoid arthritis, and the chatbot suggested a previously overlooked possibility: Hashimoto’s disease, an autoimmune disorder that attacks the thyroid gland.
Despite her doctor’s initial resistance, Bannon requested specific antibody testing. To her surprise, the results were positive. This led to further testing, which revealed two nodules in her neck. In October 2024, the diagnosis was clear: thyroid cancer. She underwent surgery in January 2025 and now remains under constant medical surveillance.
“I wouldn’t have been diagnosed without ChatGPT,” Bannon said. “I was desperate. The doctors just prescribed medication and sent me home. I knew something else was going on.” She says that if she had accepted the initial diagnosis without question, the cancer could have spread to other parts of her body.
Although Bannon acknowledges that AI isn’t a substitute for professional medical care, she encourages others to use these tools as a complement. “It doesn’t hurt to ask for more tests if something doesn’t add up. I feel lucky to be alive.”
For his part, Dr. Harvey Castro, an emergency physician in Texas and expert in medical AI, offered nuanced support for this approach. He applauds the fact that patients are becoming more actively involved in their health, but cautions: “AI can provide alerts or insights, but it’s no substitute for a trained physician.”
A similar case was reported by an X user who credited ChatGPT with detecting an urgent health issue, claiming that he would have lost an organ if he had arrived 30 minutes later. You can read that story here.