Man Poisoned Himself With ChatGPT Diet Advice

Login to Continue Learning

As ChatGPT continues to evolve, many are relying on it for various aspects of their lives, including personal health matters. Recent cases highlight the risks of such reliance; for instance, marriages have ended due to evidence allegedly uncovered by ChatGPT, and some people even use it for therapy, which has been warned against by Sam Altman. A new case involving ChatGPT’s dietary advice has emerged: a 60-year-old man in New York developed a rare and dangerous condition after following the model’s health guidance.

The Hidden Dangers of AI Health Tips: Bromism and the Salt Alternative That Went Too Far

A recent warning from a U.S. medical journal highlights the risks of seeking medical advice from ChatGPT after a man fell seriously ill following dietary advice provided by the chatbot, as reported by NBC News. This case was published in The Annals of Internal Medicine. The man developed bromism, or bromide poisoning, after replacing common table salt with sodium bromide, which he obtained online based on ChatGPT’s advice.

For three months, the man consumed sodium bromide daily, believing it to be a healthier alternative. As a result, he experienced severe symptoms including paranoia, insomnia, psychosis, and delusions that led him to believe he was being poisoned by his neighbors. He was hospitalized where doctors diagnosed bromism and provided treatment; his symptoms began to resolve after stopping the consumption of bromide.

In a unique start to his ruling against President Donald Trump’s administration on September 30, U.S. District Court Judge William Young included a scanned handwritten note sent to his office. The note read: “Trump has pardons and tanks – what do you have?” At the top of Young’s opinion in AAUP v. Rubio, which ruled that Trump’s effort to deport foreign-born student protesters was unconstitutional.

October 8, 2025

This case serves as a reminder about the dangers of purchasing substances online without oversight. It also underscores the need for caution when using AI in health contexts, as ChatGPT did not warn against the toxicity of sodium bromide. Researchers found that ChatGPT suggested bromide as an alternative to chloride without any mention of potential toxicity.

AI models should include better safety measures, particularly for sensitive topics like health advice. While we live in an age where AI tools are prevalent, curiosity should not replace caution and should never supersede professional medical advice.