Voice actor alleges a hit drama cloned his voice with AI
AI voice cloning is colliding with performers' rights again — this time over a drug ad tied to a popular Chinese series.
A working voice actor says a hit Chinese drama used an AI-cloned version of his voice in a drug advertisement and a side-story, without ever asking him. Ma Zhengyang (马正阳), a professional voice actor, posted a statement on Weibo saying he had never authorized any form of use of his voice and demanding the infringing party stop. The clip in question — a pharmaceutical ad and a "番外" (spin-off) segment attached to a currently airing drama — carried a voice that listeners and Ma himself say is a close match to his own timbre and delivery.
The production side's answer is the part that should worry every performer. When the claim surfaced, the advertising production company behind the spot said the voice was AI-generated — an admission that the sound was synthesized rather than performed, but one made after the fact rather than with the actor's consent. Jimeng AI (即梦AI), the generative-video and voice tool named in the reporting, told customers through its support channel that it needed to verify the situation. Neither response addresses the core question: who gets to approve a performer's voice being recreated, and what happens when a studio decides permission is optional?
Why it matters: this is the consent problem landing in the mainstream, not the lab. Voice cloning has been "good enough" to fool listeners for a while, but most of the public fights have been about celebrities, dead singers, or scam calls. A working jobbing actor finding his likeness used in a commercial tied to a hit show is the version that hits the everyday workforce — the people who voice ads, games, and side characters and who have the least leverage to litigate. China has been rolling out deepfake and AI-generated-content labeling rules, but those govern disclosure, not whether a voice can be cloned in the first place. A label on a synthetic ad does nothing for the actor who was never paid or asked.
The take: tools that let a studio mint a believable voice in minutes are now cheap enough to use casually, and casual use is exactly where rights erode. Until voice actors have a verifiable opt-in — a consent gate that a model provider actually enforces — "we generated it with AI" will keep functioning as a retroactive excuse rather than a defense.
What to watch: whether Ma's case pushes a performer union or platform to demand provenance checks before a synthetic voice ships.
Should AI voice tools require explicit, revocable consent from the performers they imitate — or is a disclosure label enough? Tell us in the comments.
