Skip to main content
NOPE
Low reported severity Lawsuit Filed

Shanghai: an appeal court upheld a 50,000 yuan award to a voice actor after finding that AI-synthesised audio in a company's online promotion was identifiable as the voice actor's voice

Chinese outlets reported on 29 September 2026, citing the Shanghai courts, that the Shanghai No. 1 Intermediate People's Court had concluded what it called Shanghai's first voice-rights case arising from AI-synthesised speech. According to the court's account, a voice actor who publishes work on an app learned from a friend in January 2025 that a company appeared to be using the voice actor's voice in an online user-acquisition promotion, judged the audio to be AI-synthesised, had it notarised and sued for 300,000 yuan. The company said the audio was generated by AI it had developed but that the training material could not be confirmed, and denied using the voice actor's voice. A forensic comparison rated about 90 percent of the measured features as fairly or highly similar. The first-instance court awarded 50,000 yuan and both sides appealed. The appeal court held that the company had put an identifiable synthetic voice to commercial use without consent, dismissed the appeals and upheld the award.

AI system
Unidentified voice-cloning tool
Occurred
Jan 2025
Reported
29 September 2026
Event location
Unknown
What the AI did
Depicted or impersonated the person
Reported harm
Financial Loss
Whose AI use
Someone else’s AI use
Setting
Work · Privacy · Justice
Evidence
AI involvement supported · Causal attribution supported · 3 sources, 1 underlying account
8 claims: 8 reported. 5 open questions
People reported harmed
1 person

AI system as recorded: AI speech synthesis developed by the defendant company (anonymised as Company A in the court account), which the courts found had generated audio imitating the voice actor's voice for an online promotion

What Happened

The use. According to the Shanghai courts' account, relayed by The Paper (carried on ifeng), Southern Metropolis Daily N Video (on Tencent News) and Legal Daily (carried on Sina Finance) on 29 September 2026, the claimant is a voice actor with a verified account on an app, where the claimant dubs and publishes work for pay. In January 2025 the claimant learned from a friend that a company, anonymised as Company A, appeared to be using the claimant's voice in an online user-acquisition promotion. The claimant judged the audio to be AI-synthesised, had it notarised and sued, saying there had been no cooperation or authorisation and asking for an end to the use and 300,000 yuan.

The defence. Company A said the audio was generated by AI the company had developed, but that its source and the material fed to the system could not be confirmed because a former employee had left. It said it had not known of the claimant and had not collected or used the claimant's voice.

The evidence. A forensic comparison of the notarised audio with a sample recorded from the claimant found that 24 of 28 formant indicators deviated by less than 10 percent, and that the parts rated fairly or highly similar reached 90 percent.

The rulings. The first-instance court found AI-generated voice infringement that caused the claimant economic loss and awarded 50,000 yuan. Both sides appealed. The Shanghai No. 1 Intermediate People's Court held that using a person's voice as training material without consent to generate an identifiable synthetic voice infringes that person's voice rights, and that Company A, as the party controlling the training data, had failed to prove that its material was lawfully sourced. It treated 50,000 yuan as fair, noting that the audio was online for a short time and was a process prompt sound in the promotion. The appeals were dismissed.

Limits. The three reports relay the court's own release; the judgments were not inspected. The parties are anonymised and the company and its product are not named. The date the promotion began is not reported; January 2025 is when the claimant learned of it. The Legal Daily copy on Sina Finance gives a different anonymised surname for the claimant in one sentence.

Reported harm

According to the court's account as relayed by Chinese outlets, the Shanghai courts found that AI-synthesised audio used in a company's online promotion was identifiable as a voice actor's voice and had been used without consent, causing economic loss, and awarded 50,000 yuan.

Outcome

Resolved

The first-instance court (Changning District People's Court, per the Legal Daily report carried on Sina Finance) ordered the company to pay 50,000 yuan. Both sides appealed. On 29 September 2026 the Shanghai No. 1 Intermediate People's Court dismissed the appeals and upheld the judgment (The Paper via ifeng; Southern Metropolis Daily N Video via Tencent News).

What remains unknown

  • Which company and which speech-synthesis product were involved; the court account anonymises both.
  • When the promotion began and how long the audio was in use.
  • Whether the company used the voice actor's recordings as training material; the court presumed it because the company did not prove otherwise.
  • The dates of the filing and of the first-instance judgment.
  • Where the voice actor lives.

What the evidence supports

AI involvement: supported. In the court account, the company said the audio was generated by AI it had developed, and both courts found AI-generated voice infringement. The audio imitated the voice actor's voice (depicted_or_impersonated). The company denied using the voice actor's recordings; the appeal court relied on a forensic similarity report and on the company's failure to prove the source of its training material.

8 claims: 8 reported. What the statuses mean

Reported On 29 September 2026 the Shanghai No. 1 Intermediate People's Court concluded a voice-rights case arising from AI-synthesised speech, described as the first in Shanghai, holding that the platform operator infringed and must pay 50,000 yuan for economic loss; the appeals were dismissed and the first-instance judgment upheld.

Causal attribution. Court outcome as relayed by outlets citing the court; the judgment was not inspected.

Reported In January 2025 the voice actor learned from a friend that Company A appeared to be using the voice actor's voice in an online user-acquisition promotion, judged the audio to be AI-synthesised and had it notarised before suing.

Causal attribution. The claimant's case as summarised in the court account.

  • news.ifeng.com(opens in new tab) supports · Chinese
    '2025年1月'; '从好友处了解到'; 'A公司在其互联网推广拉新活动中疑似使用了'; '该活动使用的自己声音疑似AI合成，于是对相关音频进行了公证以固定证据'
Reported The voice actor said there had been no cooperation with or authorisation of Company A and asked the court to stop the use and award 300,000 yuan.

Causal attribution. The claimant's position as summarised in the court account.

  • news.ifeng.com(opens in new tab) supports · Chinese
    '其从未与A公司进行过合作和任何授权，A公司擅自使用其声音构成侵权，遂要求A公司停止侵权并赔偿经济损失30万元'
Reported Company A said the audio was generated by AI the company had developed, but that its source and the material fed to the system could not be confirmed because a former employee had left.

Causal attribution. The company's defence as summarised in the court account; the company denied using the voice actor's voice.

  • news.ifeng.com(opens in new tab) supports · Chinese
    'A公司辩称，案涉音频确为该公司开发的AI生成，但具体的来源和投喂的素材因前员工离职无法确认'
Reported A forensic comparison found that 24 of 28 formant indicators deviated by less than 10 percent and that the parts rated fairly or highly similar reached 90 percent.

Causal attribution. Forensic opinion as summarised in the court account.

  • news.ifeng.com(opens in new tab) supports · Chinese
    '在28个共振峰声学指标中'; '有24个偏离度小于10%，其中16个小于5.36%，两者较为相似及高度相似的部分达到90%'
Reported The first-instance court awarded 50,000 yuan and both sides appealed; the Legal Daily report carried on Sina Finance says the claim was filed in the Changning District People's Court.

Causal attribution. Procedural history as relayed from the court account.

Reported The appeal court held that using a person's voice as training material without consent to generate an identifiable synthetic voice infringes that person's voice rights, and that Company A bore the consequence of failing to prove the lawful source of its material.

Causal attribution. The appeal court's reasoning as relayed; it rests on a presumption from similarity and access, because the company did not produce its training records.

  • news.ifeng.com(opens in new tab) supports · Chinese
    '未经自然人同意，使用自然人的声音作为训练语料，模仿该自然人的音色、语调和发音风格等生成能够识别该自然人的合成人声，应当认定侵害自然人声音权益'; '应承担举证不能的不利后果'
Reported The appeal court noted that the audio was online for a short time and was a process prompt sound in the promotion.

Causal attribution. The appeal court's reasoning on the amount, as relayed.

Sources

3 sources inspected, from 1 underlying account. Sources that repeat one account do not corroborate each other.

How the sources were read, and where the events happened

The Paper's article as carried on ifeng, read in the original Chinese on 2026-10-05 (HTTP 200). English renderings in this record are the research agent's (an AI's) own translations; no human translator reviewed them. Relays the Shanghai High People's Court WeChat account; the judgment itself was not inspected. Applies to s1.

Southern Metropolis Daily N Video on Tencent News, read in Chinese on 2026-10-05 (HTTP 200). English renderings in this record are the research agent's (an AI's) own translations; no human translator reviewed them. The reporter cites the Shanghai No. 1 Intermediate People's Court; same court account as s1. Applies to s2.

Legal Daily article carried on Sina Finance, read in the original Chinese on 2026-10-05 (HTTP 200). English renderings in this record are the research agent's (an AI's) own translations; no human translator reviewed them. Same court account with the first-instance court named and a judge's commentary. Applies to s3.

Event countries: Unknown. Affected-person countries: Unknown. Court countries: China.

The case was decided by the Changning District People's Court and the Shanghai No. 1 Intermediate People's Court in China. The reports anonymise the company and the app and do not say where the company is based, where its online promotion ran or where the voice actor lives, so no event or affected-person country is recorded.

Reviewed for publication 2026-10-05: Published as a core case (depicted_or_impersonated): the Shanghai courts found that a company imitated a voice actor's voice with AI-synthesised audio in a promotion without consent and awarded 50,000 yuan. One reporting chain (the court's release, read in three Chinese outlets); the parties are anonymised by the court.

People described

A voice actor with a verified account on an app who publishes dubbing work there for pay (anonymised in the court account)

People reported harmed in this case

1 person

0 AI participants · 1 other person harmed

One person, the voice actor who brought the claim (court account relayed by The Paper, Southern Metropolis Daily and Legal Daily). Exact 1.

Counted once within this case. The same person may appear in other cases. This count does not establish AI causation.

Tags

#voice-cloning#voice-rights#voice-actor#court-ruling#shanghai#china

Cite this case

Compiled per our published methodology: verification statuses, sourcing standards, and corrections process.

APA

NOPE. (2026). Shanghai: an appeal court upheld a 50,000 yuan award to a voice actor after finding that AI-synthesised audio in a company's online promotion was identifiable as the voice actor's voice. AI incidents. https://nope.net/incidents/2025-shanghai-voice-actor-ai-synthesised-voice-company-promotion-appeal-ruling

BibTeX

@misc{2025_shanghai_voice_actor_ai_synthesised_voice_company_promotion_appeal_ruling,
  title = {Shanghai: an appeal court upheld a 50,000 yuan award to a voice actor after finding that AI-synthesised audio in a company's online promotion was identifiable as the voice actor's voice},
  author = {NOPE},
  year = {2026},
  howpublished = {AI incidents},
  url = {https://nope.net/incidents/2025-shanghai-voice-actor-ai-synthesised-voice-company-promotion-appeal-ruling}
}

Related cases

High Unidentified voice-cloning tool

Khulna: a businessman paid Tk3.5 crore for old jute-mill machinery after WhatsApp calls in a voice presented as National Parliament Whip Raqibul Islam Bakul's, which the complaint and Khulna DB police describe as cloned with AI; four to five people arrested, the prime accused confessed

Khulna Gazette, the Daily Times of Bangladesh, Newsbangla24 and BD Today (all 29 September 2026) report from the first information report and Khulna Metropolitan Police Detective Branch (DB) statements that a businessman in Khulna city was joined on 29 July 2026 to a WhatsApp group call with a person introduced as National Parliament Whip Raqibul Islam Bakul, who discussed the sale of old machinery from Platinum Jute Mill and asked for advance payment. The businessman handed over Tk2 crore in cash that afternoon to men sent as representatives and a further Tk1.5 crore after inspecting the mill (30 July per Khulna Gazette and BD Today; 2 August per the Daily Times of Bangladesh and Newsbangla24), a total of Tk3.5 crore. When the businessman met the Whip in person in Khulna on 1 September, the Whip disclaimed any knowledge of it. The complaint filed on 18 September under the Cyber Security Act says the caller's voice had been changed with digital technology and artificial intelligence to pass as the Whip's; the DB deputy commissioner described the case to Asia Post as fraud by imitating the Whip's voice through AI. A press release from the Khulna city BNP media cell, sent on 4 September and published by Jaijaidin on 5 September, had already said a ring was using a US number and AI to clone the Whip's voice and demand money. Police arrested four people, recovered Tk12 lakh, and the prime accused gave a confessional statement before a magistrate on 28 September; on 29 September a court added two more detained men to the case. The voice-cloning tool is not identified and no forensic finding has been reported.

Medium Unidentified voice-cloning tool

Italy: Fideuram chairman Paolo Molesini is deceived by a WhatsApp impersonation of Intesa Sanpaolo's CEO and a phone call in the reportedly AI-cloned voice of a lawyer he knew into ordering about 95 million euros of transfers; at least 36 million remain missing and he resigned on 12 March 2026, weeks later

On 23 February 2026 Paolo Molesini, then chairman of Fideuram (the private-banking subsidiary of Intesa Sanpaolo), received a WhatsApp message from an unknown number from someone claiming to be Intesa's chief executive Carlo Messina, announcing a confidential acquisition that had to be executed through Fideuram. The same day a caller presenting as a lawyer of A&O Shearman whom Molesini knew, whose voice ANSA, Il Fatto Quotidiano and Corriere della Sera, reporting from the Milan court papers, describe as created with artificial intelligence (today.it, which also cites the seizure decree, writes that the court papers do not yet answer whether the voice was imitated), had him sign a confidentiality agreement and sent eleven payment instructions; Fideuram's treasury head was separately contacted by someone posing as Fideuram's CEO. Between 23 and 25 February eleven transfers totalling about 95 million euros went to accounts in Portugal and at Bank of China; the bank's alarm systems and the Milan prosecutors recovered about 40-42 million from China and 13 million seized in Portugal, leaving at least 36 million (39.5 million per the court papers) missing after conversion into cryptocurrency. Molesini, who Corriere writes is not under investigation, resigned as chairman on 12 March 2026, which Fideuram announced as being for personal reasons; a 48-year-old Israeli citizen is under investigation as a member of the gang.

Low Unidentified voice-cloning tool (suspected)

Game and coffee store in Havre reports phone scam in which an employee heard what sounded like the owner's voice

In a 4 February 2025 report by Hi-Line Today, the owner of a game and coffee store in a Havre shopping mall said an employee received a call on a personal phone that appeared to come from the employee's own number and carried what the employee described as the owner's voice. The owner said the scammers had also duplicated the owner's number. The article says the store reported that the scam compromised bank details and caused monetary losses; the owner said the employee was not blamed. The reporter describes the scam as using AI voice replication. The only support for a cloned voice is the employee's claim, relayed by the owner.

Low Unidentified voice-cloning tool (suspected)

Bhopal: a security guard lost about Rs 35,000 to a caller whose voice sounded like a friend, reported as a suspected AI voice-cloning fraud

Free Press Journal reported on 4 October 2026 that a 50-year-old security guard from the TT Nagar area of Bhopal, Madhya Pradesh, was cheated of nearly Rs 35,000 after a caller impersonated a friend by mimicking the friend's voice. According to the report, the call came on 30 September from an unidentified person who claimed to need money urgently and sent a QR code. The guard made three transfers (Rs 5,000, Rs 10,000 and Rs 20,000), then reached the friend on the friend's own number and learned that the friend had not called. The outlet calls it a suspected case of AI-enabled fraud with a suspected AI-cloned voice. TT Nagar police registered a case and opened an investigation. No source confirms that the voice was AI-generated.

If you or someone you know is struggling, free and confidential support is available. Find a helpline near you at Signpost.

Back to all incidents