• Researchers from the University of Chicago and the University of California, San Diego have discovered several security vulnerabilities in Microsoft's AI-powered healthcare chatbots. These vulnerabilities could potentially allow attackers to access sensitive patient information, impersonate healthcare providers, and even manipulate the chatbots' responses to provide inaccurate medical advice. The researchers have reported their findings to Microsoft, who has acknowledged the issues and is working on addressing them.

• The vulnerabilities were found in Microsoft's Healthcare Bot service, which is used by healthcare organizations to create AI-powered chatbots for patient interactions. The researchers were able to bypass the chatbots' authentication mechanisms, allowing them to impersonate authorized users and access patient data. They also discovered that the chatbots could be manipulated to provide false or misleading medical information, potentially putting patients at risk.

• The discovery of these security flaws highlights the importance of thorough security testing and oversight when it comes to AI-powered healthcare systems. As the use of AI in healthcare continues to grow, it is critical that developers and providers prioritize the protection of sensitive patient data and the integrity of medical advice. The researchers have urged Microsoft and other AI healthcare providers to address these vulnerabilities and implement robust security measures to ensure the safety and privacy of their users.