A new bill, the Healthy Technology Act of 2025, could allow artificial intelligence (AI) to prescribe medication. If the bill passes, AI could be recognized as a qualified practitioner to prescribe drugs, but only if approved by a state and the FDA.
Right now, AI is being used in healthcare, but mostly for tasks like helping doctors make decisions or managing patient records. AI tools are also being developed to help predict which medications could work best for a patient by analyzing data like electronic health records. But experts say there’s still a lot of work to be done before AI could take on the job of writing prescriptions by itself.
Can Ai write prescriptions yet? Not yet. Some AI tools are being tested to help doctors make better prescribing decisions, but they still rely on human oversight. AI can analyze lots of data to predict how a patient might react to certain treatments. But there’s no proof that AI can match or even surpass doctors when it comes to prescribing medications.
One problem is that AI can make mistakes. For example, AI could misinterpret symptoms or suggest the wrong treatment. Unlike doctors, AI can’t perform physical exams or understand a patient’s feelings about a treatment. There’s also the issue of bias, as AI is trained on human data, which may not always be complete or fair.
The Healthy Technology Act has raised questions about whether AI is ready to take on this responsibility. AI-based tools can help make decisions faster, but they still have a long way to go. Experts worry that rushing AI into prescribing could cause harm, just like in the case of self-driving cars, which are still being tested for safety. The FDA usually approves AI tools based on limited data, which isn’t enough to be sure the AI is making the right decisions for everyone. And because AI systems can change over time, it’s hard to know if they will always be accurate.
It’s possible, but it’s still a long way off for AI prescribing. Experts suggest that AI prescribing could work for simple, low-risk conditions, where a human doctor would still review the AI’s recommendations. This could help reduce errors, like when a doctor doesn’t have access to a patient’s full health record. But before AI can prescribe safely, it will need to be tested in real-world trials and prove it’s as good as or better than human doctors. For now, experts agree that AI should be a tool to help doctors, not replace them entirely. It’s a promising idea, but we need to make sure it’s ready before giving it too much power.
