Currency
  • Loading...
Weather
  • Loading...
Air Quality (AQI)
  • Loading...

AI chatbots are increasingly becoming confidants for users who share fears, hopes, family details and sensitive health symptoms. The trend raises urgent questions about the quality and safety of chatbot advice, the limits of medical guidance and who can see deeply personal conversations. According to reporting by Wirecutter, a New York Times publication, the companies behind popular assistants can access chats, and it is not always clear how that data is used.

Key Takeaways: A DuckDuckGo survey of about 2,000 Americans published in August 2026 found that 32% of chatbot users had told AI something they hid from those closest to them. OpenAI said at the start of 2026 that more than 230 million people ask ChatGPT health-related questions every week. Privacy advocates say users need clearer controls and stronger warnings before sharing sensitive information.

The appeal is simple: bots answer instantly and can seem more nonjudgmental than people around the user. Max Eddy, the article author, tested the chatbots mentioned and found many left conversations open and prompted users to disclose more. “Such interaction feels intimate immediately, sometimes alarmingly so. It does not come without consequences,” he wrote. This matters because what begins as convenience can become a permanent data trail held by companies and potentially used for training, moderation, product improvement or other purposes.

The most popular bots include Anthropic’s Claude, Google’s Gemini and OpenAI’s ChatGPT. They often indirectly or explicitly ask for personal details to complete tasks; a travel plan may require the number of travelers, whether children are coming and the preferred airport. All three companies include warnings in their documents telling users not to share personal data. Google’s warning is especially explicit: users are asked not to include confidential information they would not want a reviewer to see or used to improve Google services.

Users should assume chats may be stored, reviewed or reused unless settings prove otherwise. They can avoid names, locations, medical identifiers and other sensitive details, check data controls, use temporary chat options where available and consult clinicians for medical advice. Companies and regulators face pressure to explain retention periods, training use and third-party access. The next phase will likely focus on consent, transparency and whether health-adjacent advice should carry stronger safeguards.

Source: kun.uz