ElevenLabs Safety Policy

This document is ElevenLabs' Responsible AI framework, a publicly published governance and policy disclosure that establishes the company's stated principles, prohibited use categories, safety mechanisms, and accountability commitments for its AI voice synthesis and audio generation platform. The document states commitments to content moderation, including prohibitions on generating content that depicts minors in sexual contexts, facilitates violence, enables fraud or impersonation without consent, or produces non-consensual intimate audio; the terms also authorize platform-level enforcement including account suspension and content removal for violations of these prohibited use categories. Notably, the framework addresses voice cloning and consent requirements, stating that users may not clone or replicate another person's voice without authorization, a provision that engages emerging voice biometric protection statutes and intersects with the FTC's guidance on AI impersonation; the document asserts these obligations contractually but does not specify a formal independent audit or third-party verification mechanism for enforcement claims. The framework engages the EU AI Act, which classifies certain biometric and voice synthesis applications as high-risk or prohibited; it also intersects with GDPR obligations regarding biometric data processing, CCPA provisions for sensitive personal information including voiceprint data, and FTC Act Section 5 authority over unfair or deceptive practices in AI systems. Compliance teams should note that the document's enforceability depends significantly on jurisdiction, particularly regarding voice biometric statutes in Illinois (BIPA), Texas, and Washington, and that the framework's aspirational language coexists with limited disclosure about internal governance structures, audit cadences, or independent oversight mechanisms.