LG Berlin - 2 O 202/24

Summary

Berlin Regional Court (LG Berlin) ruled that a YouTuber unlawfully used an AI-generated voice imitating a professional German actor's distinctive voice in two political satire videos without consent. The court found no valid legal basis under Article 6(1) GDPR, rejected the controller's arguments about legitimate interest and artistic freedom, and ordered payment of €4,000 as a fictitious license fee plus €1,155.80 in legal costs. The decision establishes that AI-cloned voices of identifiable individuals constitute personal data processing and require explicit consent, even when synthetically generated.

Full text

Help LG Berlin - 2 O 202/24: Difference between revisions From GDPRhub Jump to:navigation, search VisualWikitext Latest revision as of 15:11, 28 April 2026 view source Avalang (talk | contribs)51 edits Tag: submission [1.0] (No difference) Latest revision as of 15:11, 28 April 2026 LG Berlin - 2 O 202/24 Court: LG Berlin (Germany) Jurisdiction: Germany Relevant Law: Article 6(1) GDPR § 22 KUG§ 23 KUG Decided: 20.08.2025 Published: 28.08.2025 Parties: Actor (data subject) YouTuber (controller) National Case Number/Name: 2 O 202/24 European Case Law Identifier: Appeal from: Appeal to: Unknown Original Language(s): German Original Source: openJur (in German) Initial Contributor: Ava Lang A court ordered a controller to pay €4,000 for using a data subject’s AI-generated voice without consent, holding that the processing lacked a legal basis under Article 6(1) GDPR and could not rely on legitimate interest or artistic freedom under Article 85 GDPR. Contents 1 English Summary 1.1 Facts 1.2 Holding 2 Comment 3 Further Resources 4 English Machine Translation of the Decision English Summary Facts The data subject is a widely known actor and professional voice actor. The controller operated a YouTube channel with an associated online shop. The controller published two political satire videos on YouTube. The videos used an AI-generated voice that closely imitated the data subject’s voice. Viewers in the comments identified the voice as that of the data subject. The videos ended with references to the controller’s online shop, which sold merchandise linked to the channel’s political positioning. On 28 September 2023, the data subject contacted the controller and objected to the use of his voice. He requested cessation of use and reimbursement of legal costs. The controller submitted a cease-and-desist declaration but refused to pay damages, arguing that the voice was synthetic, lawfully acquired from an AI provider, and used for satire rather than advertising. The data subject claimed that the AI-generated voice constituted use of his personal voice, that the processing occurred without consent, and that it created the impression that he endorsed the videos and products. He sought compensation equivalent to his usual licensing fees and reimbursement of legal costs. Holding First, the court held that the controller unlawfully interfered with the data subject’s right to his own voice. Although the voice was AI-generated, it closely imitated a distinctive personal characteristic attributable to the data subject. The court considered that a significant part of the audience would associate the voice with the data subject, which was sufficient to establish personal attribution. Second, the court addressed the issue under data protection law and noted that the use of an AI-generated voice imitating a real person can qualify as processing of personal data. The court examined Article 6(1) GDPR and found no valid legal basis. The data subject had not given consent under Article 6(1)(a) GDPR. Contractual necessity under Article 6(1)(b) GDPR was irrelevant. The controller relied on Article 6(1)(f) GDPR, arguing legitimate interests linked to satire and freedom of expression. The court rejected this argument after balancing interests. It found that the processing primarily served the controller’s commercial interests, as the videos increased reach and indirectly promoted the online shop. The court stressed that the controller could express political opinions and satire without using an identifiable imitation of the data subject’s voice. The data subject’s rights therefore outweighed the controller’s interests. Third, the court held that no exemption applied under Article 85 GDPR. The processing was neither journalistic nor genuinely artistic in a way that would justify overriding the data subject’s rights, particularly given the commercial context and the lack of transparency about the AI-generated nature of the voice. As a consequence, the court ordered the controller to pay €4,000 as a fictitious license fee for the unauthorised use of the voice and €1,155.80 for reimbursable legal costs, plus interest. Comment Share your comments here! Further Resources Share blogs or news articles here! English Machine Translation of the Decision The decision below is a machine translation of the German original. Please refer to the German original for more details. Judgment 1. The defendant is ordered to pay the plaintiff €4,000.00 plus interest at a rate of 5 percentage points above the base interest rate from October 16, 2023; 2. The defendant is ordered to pay the plaintiff a further €1,155.80 plus interest at a rate of 5 percentage points above the base interest rate on €1,088.60 from October 16, 2023, and on a further €67.20 from November 16, 2023. 3. The remainder of the claim is dismissed. 4. The defendant shall bear the costs of the proceedings. 5. This judgment is provisionally enforceable upon the plaintiff providing security in the amount of 110% of the amount to be enforced. Facts of the Case The parties are in dispute over claims arising from the defendant's use of an AI-generated voice. The plaintiff is a German actor, voice actor, and narrator of audiobooks and radio plays. He dubs, among others, the actor ... in his films. The defendant operates a YouTube channel under the name ..., which currently has 190,000 subscribers. The defendant also operates an online shop. For the goods sold therein, reference is made to the images on pages 3 and 4 of the reply. The defendant distributed two videos on his YouTube channel, titled "..." and "...", which featured AI-generated voiceovers and dealt with the government at the time. For further details of the videos and the voice used, reference is made to Exhibits K 1 and K 2 and the attached USB drive. By letter dated September 28, 2023 (Exhibit K6), the defendant was sued for an injunction against the plaintiff for using the plaintiff's voice and was requested to reimburse the plaintiff's extrajudicial legal fees of €1,088.60 (net), based on a value in dispute of €20,000.00, by October 15, 2023. By letter dated October 31, 2023 (Exhibit K8), the plaintiff's current legal representative, acting on behalf of the plaintiff, accepted the defendant's declaration of discontinuance and unsuccessfully demanded that the defendant pay damages of €2,000.00 per clip, as well as reimburse the extrajudicial legal fees, now calculated based on a value in dispute of €24,000.00, by November 14, 2023. The plaintiff alleges that the defendant used the plaintiff's voice, which was created by the plaintiff. His voice was also identified by the video commentators as the dubbing voice of ... or his own voice. By using his voice, the defendant infringes upon his general right of personality, specifically the right to his own voice. It makes no difference whether the voice is generated by a real voice impersonator or a recording. The licensing damages of €2,000 in each case correspond to his usual fee practice. The defendant must also reimburse him for his out-of-court legal fees. The plaintiff requests: 1. that the plaintiff be paid €4,000.00 plus interest at a rate of 9% above the base interest rate from September 28, 2023; 2. that the plaintiff be paid €1,088.60 plus interest at a rate of 9% above the base interest rate from September 28, 2023; 3. to pay the plaintiff €67.20 plus interest at a rate of 5% above the base interest rate from the date the lawsuit was filed. The defendant requests that the lawsuit be dismissed. He argues that when creating his satirical videos, he simply chose an authentic voice with a heroic tone and used the synthetic imitation of a voice suggested to him by the plaintiff's software—not an original spoken voice. Rather, it is an AI-generated voice similar to the vocal range of [name omitted], but it is not the same voice—that is, not the plaintiff's voice. This was also demonstrated in the oral hearing on March 25, 2025.

Entities