Advanced audio AI technologies—including speech generation, source separation, and acoustic scene understand- ing—could drive innovations across smart devices, assisted living, smart buildings, and urban noise monitoring. But a growing body of research indicates that privacy is still an important concern for a variety of stakeholders related to AI solutions becoming part of the real-world. This concern is heightened as conventional voice privacy approaches remain vulnerable to attacks, and there is a tension between offering novel functionality and privacy protection. To address this emerging problem for AI audio technology, this panel brings together experts in cybersecurity, voice privacy, responsible AI, audio forensics, and audio AI to examine current issues and emerging challenges.