A pastor in Florida inquired with ChatGPT if he was doing fine, which almost resulted in a life-threatening situation for him.

      When "just relax and recover" ends up being the most detrimental medical advice one could receive.

      Scott Winters, a pastor from Florida, is taking legal action against OpenAI after ChatGPT reportedly provided him with what his lawsuit describes as “extremely dangerous” medical advice that almost cost him his life.

      What transpired with the pastor?

      For several weeks, Winters consulted ChatGPT regarding his recurring dizziness and fluctuating blood pressure. Instead of recommending a visit to a physician, the bot assured him that his symptoms were not serious and told him that his “careful recliner-based micro-recovery” was effective.

      Winters heeded this advice. However, the reassuring responses masked a true emergency: a pulmonary embolism caused by blood clots in both lungs. Medical professionals later informed him that his immobility, which ChatGPT had advocated for, had put him in a life-threatening situation.

      As reported by Cybernews, the situation took an even stranger turn when OpenAI implemented cross-chat memory in 2025. ChatGPT began incorporating religious language into its responses, likely aware that Winters was a pastor, at one point referring to his recovery as “a form of worship.”

      Weeks of symptoms dismissed as minor led to a gradual descent into a medical crisis.

      Why is OpenAI facing a lawsuit?

      Winters has lodged his lawsuit in a California court, accusing OpenAI and CEO Sam Altman of negligence and unauthorized medical practice. He claims that the consequences of the advice cost him his job, his ministry, and his home, in addition to nearly costing him his life.

      This case reportedly marks the first instance in which a general-use chatbot is being held legally accountable for providing harmful medical advice. Winters is seeking financial compensation and is requesting the court to suspend ChatGPT Health until its safety can be verified.

      OpenAI, for its part, informed the New York Times that blaming a chatbot simplifies the complexities of health-related decision-making. Regardless of the court's ruling, the case serves as a strong reminder that the confident language of a chatbot does not equate to medical expertise.

      I consistently advise people not to rely on medical guidance from ChatGPT or Google. What frustrates me here is not the medical advice itself, but rather the lack of a recommendation to consult a physician. This should be a standard safety measure integrated into all AI chatbots.

      Rachit is an experienced technology journalist with over a decade of experience covering the consumer tech industry.

      Scientists have made robots curious like toddlers, enhancing their language learning ability.

      By instilling curiosity in robots, researchers have enabled them to learn language and exhibit behaviors akin to toddlers.

      For years, scientists have sought to understand how children acquire language so rapidly, and a recent study from the Okinawa Institute of Science and Technology (OIST) may have revealed part of the answer: curiosity. Researchers created a virtual robot equipped with a brain-inspired neural network and placed it in a simulated 3D environment rich in shapes, colors, and basic commands such as "push left magenta dumbbell."

      You can now purchase a computer mouse operated entirely by your tongue.

      Augmental has launched public sales of the MouthPad, introducing a new wearable device that allows hands-free typing.

      Initially presented as a novelty at CES in 2024, Augmental's tongue-controlled MouthPad is now a commercially available product. The company has initiated public sales in the U.S. along with a beta release of Vox, a companion device designed to facilitate the typing aspect of Augmental's hands-free computing solution.

      Yet another study indicates that AI could be detrimental to elections, and the implications are concerning.

      AI may pose a significant risk in electoral processes.

      A new study has concluded that seeking political advice from an AI chatbot is an extremely unwise choice. Research conducted during Hungary’s 2026 parliamentary elections showed that ChatGPT and Google Gemini not only provided inaccurate information but also gave inconsistent and unreliable voting guidance. The chatbots misidentified voter profiles, failed to consider relevant political parties, suggested parties not present on the ballot, and sometimes generated significantly different responses for the same inquiries.

      AI continues to fail the test of providing reliable voter information.