SkillDetective - Automated Policy-Violation Detection of Voice Assistant Applications in the Wild

Explore the groundbreaking research presented at USENIX Security '22 on SkillDetective, an automated tool for detecting policy violations in voice assistant applications. Delve into the work of Jeffrey Young, Song Liao, Long Cheng, Hongxin Hu, and Huixing Deng from Clemson University and the University at Buffalo as they address the critical issue of policy compliance in third-party voice-apps. Learn how SkillDetective conducts comprehensive dynamic analysis to evaluate voice-apps' adherence to 52 different policy requirements, examining textual, image, and audio content. Discover the alarming results of their extensive testing, which revealed thousands of policy-violating apps across Amazon Alexa skills and Google Assistant actions. Gain insights into the importance of automated policy enforcement in the rapidly expanding world of voice personal assistant services and the potential implications for user privacy and security.