Ensuring Safety in the Generative AI Ecosystem: Protecting Users from Non-Consensual Intimate Content

Google Play outlines its policies and enforcement measures against apps that facilitate non-consensual intimate imagery (NCII), including 'nudify' and deepfake generation tools. It describes multi-layered defenses such as lifecycle testing of apps, cutting off monetization and ad pathways for violators, and partnerships with NCII-defense organizations through its Priority Flagger Program. Developers are given practical guidance for app review, including giving reviewers full access to AI features, documenting safety testing against adversarial prompts, building custom input/output moderation beyond model-native filters, monitoring ad campaigns for compliance, and using user feedback signals to strengthen guardrails over time.

Nguồn: https://android-developers.googleblog.com/2026/08/ensuring-safety-genai-preventing-non-consensual-intimate-content.html. 8 Sync News chỉ tóm tắt và dẫn link; bản quyền nội dung thuộc tác giả và nguồn gốc.