OpenAI has officially introduced ChatGPT Health, a new feature designed to help users ask health-related questions in a more private and personalized environment.
The tool appears as a separate tab inside ChatGPT and is being positioned as an AI “healthcare ally,” though the company clearly states it is not meant to diagnose or treat medical conditions.
ChatGPT Health works as a sandboxed space with its own chat history and memory, separate from regular ChatGPT conversations. OpenAI is encouraging users to connect personal medical records and popular wellness apps such as Apple Health, MyFitnessPal, Weight Watchers, Peloton, and Function. By doing this, the chatbot can offer more grounded responses based on lab results, clinical summaries, fitness activity, sleep patterns, diet tracking, and workout data.
To support medical record access, OpenAI has partnered with b.well, a healthcare platform that connects with around 2.2 million providers. This allows users to upload visit summaries, lab reports, and medical histories so ChatGPT Health can help explain test results or prepare users for doctor appointments. The feature is currently available through a waitlist as part of a beta launch, with plans to roll it out gradually to all users, including free accounts.
OpenAI says ChatGPT Health can help users understand medical reports, plan healthier diets, adjust workout routines, and even compare insurance options based on personal healthcare patterns. According to the company, more than 230 million people already ask ChatGPT health and wellness questions every week, and many of these conversations happen outside normal clinic hours, especially in underserved or rural areas.
However, AI-generated medical advice remains controversial. OpenAI acknowledges the risks and emphasizes that ChatGPT Health is not a replacement for doctors. The company points to past efforts, including working with over 260 physicians who reviewed model responses more than 600,000 times across dozens of medical topics. Despite these efforts, AI tools from multiple companies have previously been criticized for giving misleading or dangerous health advice.
Mental health was notably downplayed in OpenAI’s official announcement, even though many users already turn to ChatGPT for emotional support. During a media briefing, OpenAI said ChatGPT Health can handle mental health topics but is designed to respond carefully in distress situations and encourage users to seek professional help or support from trusted people. The company also says it has tuned the model to avoid alarmist responses that could worsen health anxiety or hypochondria.
On the privacy side, OpenAI claims ChatGPT Health operates with enhanced security protections and multiple layers of encryption, though it does not use end-to-end encryption. Health conversations are not used to train OpenAI’s main AI models by default. If a health discussion starts in regular ChatGPT, users may be prompted to move it into the Health tab for added protection. Still, OpenAI admits that data may be shared if required by law or in emergencies.
OpenAI also clarified that ChatGPT Health is not covered under HIPAA, as the law applies to clinical healthcare providers rather than consumer AI products. This distinction may raise concerns for users who plan to upload sensitive medical data.
If this article helped you, please consider supporting our work. Every small contribution keeps Abijita.com independent and running.
With ChatGPT Health, OpenAI is taking a major step into digital health, offering convenience and personalized insights while trying to balance safety, privacy, and responsibility. Whether users treat it as a helpful assistant or rely on it too heavily will likely shape the debate around AI’s role in healthcare in the months ahead.
