• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    China’s top court sets consent rules for AI face and voice replicas, a post says

    The account says victims can seek rapid injunctions when delays would cause hard-to-repair harm under guidelines it says were issued on September 7, 2026.

    BR
    1 Source, ,

    TLDR

    A post says China’s Supreme People’s Court issued its first dedicated guidelines on AI disputes on September 7, 2026. According to the account, creating or distributing a recognizable digital replica of someone’s face or voice without consent infringes personality rights. It also describes liability for using replicas to spread false or defamatory claims, especially sexual defamation, and says victims can seek rapid injunctions when waiting would cause harm that is hard to repair.

    Combined views

    19.2K

    1 Source, first seen 23d ago

    Combined views

    19.2K

    1 Source, first seen 23d ago

    179 likes
    23d ago
    first seen 23d ago
    179 likes
    32 comments
    37 saves
    39 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    32 comments
    37 saves
    39 reposts
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    1 Source

    @BrianRoemmeleIn China Your Own Your Likeness A Supreme Court Just Found. Here is the US, we will have decades of movie and record companies sending us to the Stone Age. A familiar face appears on the video call. The voice is right. The cadence is right. The slight pause before the punchline is right. It is not them. On Monday, September 7, 2026, China’s Supreme People’s Court drew a legal line through it. The top court issued its first dedicated judicial guidelines on AI disputes, making clear that creating or distributing a recognizable digital replica of another person—face or voice—without consent is an infringement of personality rights. The guidelines land at a precise historical pressure point. China is racing the United States for AI dominance while President Xi Jinping has repeatedly warned that the technology must remain “secure and controllable.” The court, through Vice President Tao Kaiyuan, framed the 24-article opinion as an attempt to “balance development and security”—let the horse run fast, but keep it from kicking the barn down. The core rule is simple and sharp: you do not get to mint someone else’s face or voice as raw material. Cloned faces, cloned voices, and the use of those replicas to spread false or defamatory claims—especially sexual defamation—are now explicit grounds for liability. Victims can seek rapid injunctions when waiting would cause harm that is hard to repair. Service providers who are notified of infringing output and fail to act promptly can be held liable. Users who deliberately prompt systems to generate harmful content can be held liable too. That last point matters. The court is not treating the model as a mysterious black box that swallows responsibility. Both the person who types the prompt and the company that serves the model sit inside the accountability chain. The opinion also reaches beyond deepfakes. Algorithmic price discrimination—“killing the familiar,” the practice of showing different prices to different customers for the same good without justification—is called out. So are AI hallucinations when they produce false information that damages rights. The court left some questions open on purpose, saying consensus has not yet formed and experience must accumulate before the next layer of rules is written. This is not China’s first move. Labeling rules for AI-generated content already took effect in September 2025. Earlier regulations targeted deepfake livestreams of public figures. What is new is a national judicial map that lower courts can actually use when a neighbor, a celebrity, or a small business owner walks in with a video that looks and sounds exactly like them saying something they never said. The technology that made the problem cheap is the same technology that made the problem urgent. A few years ago, high-quality face and voice cloning required specialist labs. Now a phone and a handful of public clips can produce a replica good enough to fool a family member on a video call. That compression of capability is why the court moved from general principles to specific red lines. The deeper story is not only legal. It is about who owns the surface of a human being once that surface can be photocopied at industrial scale. A face used to be expensive to counterfeit. A voice used to be expensive to steal. Both are now cheap. The Supreme People’s Court just announced that cheap does not mean free. Whether other jurisdictions follow the same bright-line approach—or keep treating deepfakes as a patchwork of existing privacy, defamation, and publicity laws—will shape the next decade of identity. For now, in China, the replica is no longer an unregulated gray zone. It is evidence. And the person whose face or voice was taken has a court they can walk into.

    1 Source

    @BrianRoemmeleIn China Your Own Your Likeness A Supreme Court Just Found. Here is the US, we will have decades of movie and record companies sending us to the Stone Age. A familiar face appears on the video call. The voice is right. The cadence is right. The slight pause before the punchline is right. It is not them. On Monday, September 7, 2026, China’s Supreme People’s Court drew a legal line through it. The top court issued its first dedicated judicial guidelines on AI disputes, making clear that creating or distributing a recognizable digital replica of another person—face or voice—without consent is an infringement of personality rights. The guidelines land at a precise historical pressure point. China is racing the United States for AI dominance while President Xi Jinping has repeatedly warned that the technology must remain “secure and controllable.” The court, through Vice President Tao Kaiyuan, framed the 24-article opinion as an attempt to “balance development and security”—let the horse run fast, but keep it from kicking the barn down. The core rule is simple and sharp: you do not get to mint someone else’s face or voice as raw material. Cloned faces, cloned voices, and the use of those replicas to spread false or defamatory claims—especially sexual defamation—are now explicit grounds for liability. Victims can seek rapid injunctions when waiting would cause harm that is hard to repair. Service providers who are notified of infringing output and fail to act promptly can be held liable. Users who deliberately prompt systems to generate harmful content can be held liable too. That last point matters. The court is not treating the model as a mysterious black box that swallows responsibility. Both the person who types the prompt and the company that serves the model sit inside the accountability chain. The opinion also reaches beyond deepfakes. Algorithmic price discrimination—“killing the familiar,” the practice of showing different prices to different customers for the same good without justification—is called out. So are AI hallucinations when they produce false information that damages rights. The court left some questions open on purpose, saying consensus has not yet formed and experience must accumulate before the next layer of rules is written. This is not China’s first move. Labeling rules for AI-generated content already took effect in September 2025. Earlier regulations targeted deepfake livestreams of public figures. What is new is a national judicial map that lower courts can actually use when a neighbor, a celebrity, or a small business owner walks in with a video that looks and sounds exactly like them saying something they never said. The technology that made the problem cheap is the same technology that made the problem urgent. A few years ago, high-quality face and voice cloning required specialist labs. Now a phone and a handful of public clips can produce a replica good enough to fool a family member on a video call. That compression of capability is why the court moved from general principles to specific red lines. The deeper story is not only legal. It is about who owns the surface of a human being once that surface can be photocopied at industrial scale. A face used to be expensive to counterfeit. A voice used to be expensive to steal. Both are now cheap. The Supreme People’s Court just announced that cheap does not mean free. Whether other jurisdictions follow the same bright-line approach—or keep treating deepfakes as a patchwork of existing privacy, defamation, and publicity laws—will shape the next decade of identity. For now, in China, the replica is no longer an unregulated gray zone. It is evidence. And the person whose face or voice was taken has a court they can walk into.