Particle.news

China’s Top Court Issues Rules Banning Unauthorised AI Deepfakes

Platforms will face possible legal liability for failing to act on rights‑infringing AI content, signaling a phased approach to policing algorithmic harms.

Overview

  • The Supreme People’s Court’s guidelines bar creating or sharing recognisable AI copies of real people’s faces or voices without consent, covering deepfakes and voice cloning.
  • The rules make platforms and AI service providers potentially liable if they do not take timely action after being notified that their systems produced rights‑infringing content.
  • The guidance expands legal remedies for victims, including people whose faces were digitally altered to spread false and defamatory sexual claims.
  • The court also addresses other AI disputes such as algorithmic price discrimination and AI‑generated false information and sets procedural rules for handling these cases.
  • Officials framed the move as balancing fast AI development with risk control and said some contested issues will be left for future clarification as experience and consensus grow.