Shen Anyu, a 31‑year‑old voice actor from Xuzhou, has spent the past year recording videos to prove he is not a synthetic voice. AI replicas of his voice, circulating since 2025, trigger platform filters that tag his real work as synthetic, cutting his reach and earnings.

The clones are produced by open‑source deep‑learning models that learn a speaker’s timbre from only a few seconds of audio. Platforms lack reliable detection, so Shen’s recordings are mislabelled automatically. The result is a sharp drop in client requests, with some projects now opting for a cheap AI voice that mimics his tone.

Shen’s monthly income was $1,500 in quiet months and rose to $30,000 during peak demand. After AI versions appeared, clients stopped hiring him and offered lower rates for the cloned voice, forcing Shen to file complaints, contact uploaders, and consult lawyers. Legal action is costly and rarely succeeds.

The case highlights a gap in creator protection for AI Gamer voice work. Without robust synthetic‑voice detection and clearer intellectual‑property rules, performers risk losing income and identity to automated replicas.