| Chinese Original | English Translation |
|---|---|
| 9月11日上午，北京互联网法院召开涉人工智能案件审理情况新闻发布会（点击查看），会上发布了八起典型案例。具体案例如下： | On the morning of September 11th, Beijing Internet Court held a press conference on the trial situation of artificial intelligence-related cases (click to view), where eight typical cases were released. The specific cases are as follows: |
| Chinese Original | English Translation |
|---|---|
| 案例一 | Case One |
| 李某某诉刘某某侵害作品署名权和信息网络传播权纠纷案——AI文生图的法律属性及权利归属的认定 | Li v. Liu Case of Infringement of Work Attribution Rights and Information Network Dissemination Rights - Legal Nature and Rights Attribution of AI-Generated Images |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 2023年2月24日，原告李某某为画出一幅在黄昏的光线条件下具有摄影风格的少女特写，使用生成式人工智能大模型Stable Diffusion，通过选取模型、输入提示词及反向提示词、设置生成参数等操作生成图片，并于2023年2月26日发布在小红书平台。 | On February 24, 2023, plaintiff Li, in order to create a photographic-style close-up portrait of a girl under twilight lighting conditions, used the generative artificial intelligence large model Stable Diffusion, generating images through operations such as selecting models, inputting prompts and negative prompts, and setting generation parameters, and published it on the Xiaohongshu platform on February 26, 2023. |
| 2023年3月2日，被告刘某某在其注册及使用的百家号账号发布文章，将涉案图片作为配图，配图中截去了原告的署名水印，未说明图片具体来源。据此，原告李某某提起诉讼，请求判令被告刘某某赔礼道歉并赔偿经济损失。 | On March 2, 2023, defendant Liu published an article on his registered and used Baijiahao account, using the disputed image as an illustration, removing the plaintiff's signature watermark from the illustration, without indicating the specific source of the image. Based on this, plaintiff Li filed a lawsuit, requesting the court to order defendant Liu to apologize and compensate for economic losses. |
| 经当庭勘验，在使用生成式人工智能大模型Stable Diffusion时，通过变更个别提示词或者变更个别参数，大模型生成的图片不同。 | Through courtroom examination, when using the generative artificial intelligence large model Stable Diffusion, changing individual prompt words or individual parameters results in different images generated by the large model. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，从涉案图片的外观上来看，涉案图片与通常人们见到的照片、绘画无异，显然属于艺术领域，且具有一定的表现形式。 | The court held that from the appearance of the disputed image, it is no different from photographs and paintings that people usually see, clearly belonging to the artistic field and having certain forms of expression. |
| 涉案图片系原告利用生成式人工智能技术生成的，从原告构思涉案图片起，到最终选定涉案图片止，原告进行了一定的智力投入，故涉案图片具备"智力成果"要件。 | The disputed image was generated by the plaintiff using generative artificial intelligence technology. From the plaintiff's conception of the disputed image to the final selection of the disputed image, the plaintiff made certain intellectual investments, so the disputed image meets the "intellectual achievement" requirement. |
| 从涉案图片本身来看，体现出了与在先作品存在可以识别的差异性，体现了原告的选择和安排，调整修正过程亦体现了原告的审美选择和个性判断。 | From the disputed image itself, it shows identifiable differences from prior works, reflecting the plaintiff's choices and arrangements, and the adjustment and correction process also reflects the plaintiff's aesthetic choices and individual judgment. |
| 在无相反证据的情况下，可以认定涉案图片由原告独立完成，体现出了原告的个性化表达，故涉案图片具备"独创性"要件。 | In the absence of contrary evidence, it can be determined that the disputed image was independently completed by the plaintiff and reflects the plaintiff's personalized expression, so the disputed image meets the "originality" requirement. |
| 涉案图片是以线条、色彩构成的有审美意义的平面造型艺术作品，属于美术作品，受到著作权法的保护。 | The disputed image is a planar visual art work composed of lines and colors with aesthetic significance, belonging to fine art works and protected by copyright law. |
| 就涉案作品的权利归属而言，著作权法规定，作者限于自然人、法人或非法人组织，因此人工智能模型本身无法成为我国著作权法上的作者。 | Regarding the rights attribution of the disputed work, copyright law stipulates that authors are limited to natural persons, legal persons, or unincorporated organizations, so artificial intelligence models themselves cannot become authors under China's copyright law. |
| 原告根据需要对涉案人工智能模型进行相关设置，并最终选定涉案图片的人，涉案图片是基于原告的智力投入直接产生，且体现出原告的个性化表达，故原告是涉案图片的作者，享有涉案图片的著作权。 | The plaintiff made relevant settings to the disputed artificial intelligence model according to needs and was the person who finally selected the disputed image. The disputed image was directly produced based on the plaintiff's intellectual investment and reflects the plaintiff's personalized expression, so the plaintiff is the author of the disputed image and enjoys the copyright of the disputed image. |
| 被告未经许可，使用涉案图片作为配图并发布在自己的账号中，使公众可以在其选定的时间和地点获得涉案图片，侵害了原告就涉案图片享有的信息网络传播权。 | The defendant, without permission, used the disputed image as an illustration and published it on his own account, enabling the public to obtain the disputed image at times and places of their choosing, infringing the plaintiff's right of information network dissemination regarding the disputed image. |
| 此外，被告将涉案图片进行去除署名水印的处理，侵害了原告的署名权，应当承担侵权责任。 | In addition, the defendant processed the disputed image by removing the signature watermark, infringing the plaintiff's right of attribution and should bear tort liability. |
| 判决被告刘某某赔礼道歉并赔偿经济损失。宣判后，双方均未提起上诉，该判决已生效。 | The court ruled that defendant Liu should apologize and compensate for economic losses. After the judgment, neither party appealed, and the judgment has taken effect. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 本案明确人利用人工智能生成的内容，若符合作品的定义，则应被认定为作品，受到著作权法保护。 | This case clarifies that content generated by humans using artificial intelligence, if it meets the definition of a work, should be recognized as a work and protected by copyright law. |
| 同时人工智能生成内容的著作权权属可根据开发者、使用者、所有者等各参与方对人工智能生成内容的独创性智力贡献进行认定。 | At the same time, the copyright ownership of AI-generated content can be determined based on the original intellectual contributions of various parties such as developers, users, and owners to the AI-generated content. |
| 本案在回应技术变革带来权利认定新问题的同时，为人工智能生成内容的司法审查提供了制度适应与规则回应的实践范式，具有引领性和方向性意义。 | While responding to new issues in rights determination brought by technological transformation, this case provides a practical paradigm of institutional adaptation and rule response for judicial review of AI-generated content, having leading and directional significance. |
| 本案入选新时代推动法治进程2024年度十大提名案件，2024中国数字经济发展与法治建设十个重大影响力事件、2023年度中国法治实施十大事件等。 | This case was selected as one of the top ten nominated cases for promoting the rule of law process in the new era in 2024, one of the ten major influential events in China's digital economy development and rule of law construction in 2024, and one of the top ten events in China's rule of law implementation in 2023. |
| Chinese Original | English Translation |
|---|---|
| 案例二 | Case Two |
| 殷某某诉某智能科技公司等人格权侵权纠纷案——自然人声音权益可及于AI生成声音 | Yin v. An Intelligent Technology Company and Others Personality Rights Infringement Dispute - Natural Person's Voice Rights Can Extend to AI-Generated Voices |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 原告殷某某是一名配音演员，发现他人利用其配音制作的作品在多个知名APP广泛流传。经声音筛选和溯源，发现上述作品中的声音来自于被告一某智能科技公司运营的平台中的文本转语音产品。 | Plaintiff Yin is a voice actor who discovered that works created using his dubbing were widely circulated on multiple well-known APPs. Through voice screening and tracing, it was found that the voices in the above works came from text-to-speech products on a platform operated by defendant No. 1, an intelligent technology company. |
| 原告曾接受被告二某文化传媒公司的委托录制录音制品，被告二为录音制品的著作权人。后被告二将原告为其录制的录音制品的音频提供给被告三某软件公司。 | The plaintiff had previously accepted a commission from defendant No. 2, a cultural media company, to record audio products, with defendant No. 2 being the copyright owner of the recordings. Later, defendant No. 2 provided the audio of the recordings made by the plaintiff to defendant No. 3, a software company. |
| 被告三仅以原告录制的一部录音制品作为素材进行AI化处理，生成了涉案文本转语音产品并在被告四某网络科技公司运营的云服务平台对外出售。 | Defendant No. 3 used only one recording made by the plaintiff as material for AI processing, generated the disputed text-to-speech product, and sold it externally on a cloud service platform operated by defendant No. 4, a network technology company. |
| 被告一某智能科技公司与被告五某科技发展公司签订在线服务买卖合同，由被告五向被告三下单采购，其中包括了涉案文本转语音产品。 | Defendant No. 1, the intelligent technology company, signed an online service sales contract with defendant No. 5, a technology development company, with defendant No. 5 placing orders with defendant No. 3 for procurement, including the disputed text-to-speech product. |
| 被告一某智能科技公司采取应用程序接口形式，在未经技术处理的情况下，直接调取并生成文本转语音产品在其平台中使用。 | Defendant No. 1, the intelligent technology company, used application program interface forms to directly call and generate text-to-speech products for use on its platform without technical processing. |
| 原告主张，被告的行为已经严重侵犯了原告的声音权益，被告一某智能科技公司、被告三某软件公司应立即停止侵权、赔礼道歉，五被告应当赔偿原告经济损失、精神损失。 | The plaintiff claimed that the defendants' behavior had seriously infringed upon the plaintiff's voice rights, and defendants No. 1 and No. 3 should immediately stop the infringement and apologize, while all five defendants should compensate the plaintiff for economic and mental damages. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，自然人声音以声纹、音色、频率为区分，具有独特性、唯一性、稳定性特点，能够给他人形成或引起一般人产生与该自然人有关的思想或感情活动，可以对外展示个人的行为和身份。 | The court held that natural person's voice is distinguished by voiceprint, timbre, and frequency, having characteristics of uniqueness, singularity, and stability, capable of forming or causing others to generate thoughts or emotional activities related to that natural person, and can display personal behavior and identity externally. |
| 自然人声音的可识别性是指在他人反复多次或长期聆听的基础上，通过该声音特征能识别出特定自然人。 | The recognizability of a natural person's voice refers to the ability to identify a specific natural person through voice characteristics based on others' repeated or long-term listening. |
| 利用人工智能合成的声音，如果能使一般社会公众或者相关领域的公众根据其音色、语调和发音风格，关联到该自然人，可以认定为具有可识别性。 | Voices synthesized using artificial intelligence, if they can enable the general public or relevant field public to associate them with a natural person based on timbre, intonation, and pronunciation style, can be determined to have recognizability. |
| 本案中，因被告三系仅使用原告个人声音开发涉案文本转语音产品，而且经当庭勘验，该AI声音与原告的音色、语调、发音风格等具有高度一致性，能够引起一般人产生与原告有关的思想或感情活动，能够将该声音联系到原告本人，进而识别出原告的主体身份。 | In this case, because defendant No. 3 developed the disputed text-to-speech product using only the plaintiff's personal voice, and through courtroom examination, the AI voice has high consistency with the plaintiff's timbre, intonation, pronunciation style, etc., capable of causing ordinary people to generate thoughts or emotional activities related to the plaintiff, able to connect the voice to the plaintiff himself, and thereby identify the plaintiff's subject identity. |
| 被告二对录音制品享有著作权等权利，但不包括授权他人对原告声音进行AI化使用的权利。 | Defendant No. 2 enjoys copyright and other rights to the recording, but this does not include the right to authorize others to use the plaintiff's voice for AI purposes. |
| 被告二与被告三公司签订数据协议，在未经原告本人知情同意的情况下，授权被告三AI化使用原告声音的行为无合法权利来源。 | Defendant No. 2 signed a data agreement with defendant No. 3 company, authorizing defendant No. 3 to use the plaintiff's voice for AI purposes without the plaintiff's knowledge and consent, which has no legitimate source of rights. |
| 因此，被告二、被告三关于获得原告合法授权的抗辩不能成立。 | Therefore, defendants No. 2 and No. 3's defense regarding obtaining legitimate authorization from the plaintiff cannot be established. |
| 被告二、被告三未经原告许可AI化使用了原告声音，构成对原告声音权益的侵犯，其侵权行为造成了原告声音权益受损的后果，应承担相应的法律责任。 | Defendants No. 2 and No. 3 used the plaintiff's voice for AI purposes without the plaintiff's permission, constituting infringement of the plaintiff's voice rights, and their infringing acts caused damage to the plaintiff's voice rights, so they should bear corresponding legal responsibilities. |
| 被告一、被告四、被告五主观上不存在过错，不承担损害赔偿责任。 | Defendants No. 1, No. 4, and No. 5 have no subjective fault and do not bear liability for damages. |
| 因此，综合考量被告侵权情节、同类市场产品价值、产品播放量等因素，对损害赔偿予以酌定。 | Therefore, comprehensively considering factors such as the defendants' infringement circumstances, the value of similar market products, and product play volumes, damages are determined at the court's discretion. |
| 判决被告一、被告三向原告书面赔礼道歉，被告二、被告三赔偿原告经济损失。宣判后，双方均未提起上诉，该判决已生效。 | The court ruled that defendants No. 1 and No. 3 should provide written apologies to the plaintiff, and defendants No. 2 and No. 3 should compensate the plaintiff for economic losses. After the judgment, neither party appealed, and the judgment has taken effect. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 本案明确了经人工智能技术处理的声音受声音权保护的认定标准，为新业态、新技术的应用划定了行为界限，有助于规范和引导人工智能技术沿着为民、向善的方向发展。 | This case clarified the recognition standards for voice rights protection of voices processed by artificial intelligence technology, demarcated behavioral boundaries for the application of new business forms and new technologies, helping to regulate and guide the development of artificial intelligence technology in a people-oriented and virtuous direction. |
| 本案入选最高人民法院发布的民法典颁布五周年典型案例，利用网络、信息技术侵害人格权典型案例，2024年度中国十大传媒法事例。 | This case was selected as a typical case for the fifth anniversary of the Civil Code promulgation released by the Supreme People's Court, a typical case of infringing personality rights using network and information technology, and one of China's top ten media law cases in 2024. |
| Chinese Original | English Translation |
|---|---|
| 案例三 | Case Three |
| 李某某诉某文化传媒有限公司网络侵权责任纠纷案——未经权利人许可使用AI合成的名人声音用于"带货"构成侵权，委托推广商家应承担连带责任 | Li v. A Cultural Media Co., Ltd. Network Tort Liability Dispute - Using AI-Synthesized Celebrity Voices for "Live Streaming Sales" Without Permission Constitutes Infringement, Commissioning Merchants Should Bear Joint Liability |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 原告李某某在教育、育儿领域具有一定知名度和社会影响力。 | Plaintiff Li has certain popularity and social influence in the education and parenting fields. |
| 2024年，原告李某某发现被告某文化传媒有限公司在其运营的某网络平台店铺中，通过使用原告李某某的公开演讲、授课视频，并配以与原告声音高度近似的AI合成声音，对其销售多本的家庭教育类图书进行宣传推介。 | In 2024, plaintiff Li discovered that defendant, a cultural media company, used plaintiff Li's public speeches and teaching videos in its online platform store, combined with AI-synthesized voices highly similar to the plaintiff's voice, to promote and recommend multiple family education books it was selling. |
| 原告认为被告未经许可使用原告的肖像和通过AI合成的声音制作宣传产品，使原告的人格形象与其商业宣传对象形成紧密关联，从而使消费者误以为原告是其销售图书的代言人或推介者，利用原告人格形象、专业背景和社会影响力吸引关注，增加交易机会，侵犯了原告的肖像权和声音权。 | The plaintiff believed that the defendant used the plaintiff's likeness and AI-synthesized voice to create promotional products without permission, creating close associations between the plaintiff's personal image and the defendant's commercial promotion targets, causing consumers to mistakenly believe the plaintiff was an endorser or recommender of the books being sold, using the plaintiff's personal image, professional background, and social influence to attract attention and increase business opportunities, infringing the plaintiff's portrait rights and voice rights. |
| 被告作为图书销售者，与视频发布者（某带货主播）之间为委托关系，共同完成销售活动，被告对主播发布的视频具有审查义务和能力，对涉案视频的发布应当承担赔礼道歉、赔偿损失等侵权责任。 | As a book seller, the defendant has a commissioning relationship with the video publisher (a live streaming host), jointly completing sales activities. The defendant has the obligation and ability to review videos published by the host and should bear tort liability including apology and compensation for damages regarding the publication of the disputed videos. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，涉案视频使用了原告李某某的肖像以及AI合成声音，该声音与原告李某某本人的声音在音色、语调、发音风格上具有高度一致性，结合原告李某某在教育、育儿领域的知名度，涉案视频对家庭教育类书籍进行宣传推介，更易使观看涉案视频的公众将视频中的相关内容与原告李某某之间建立联系，可以认定一定范围内的听众能够将涉案AI合成声音与原告本人建立一一对应的联系。 | The court held that the disputed video used plaintiff Li's likeness and AI-synthesized voice, which has high consistency with plaintiff Li's actual voice in timbre, intonation, and pronunciation style. Combined with plaintiff Li's popularity in education and parenting fields, the disputed video promoted family education books, making it easier for viewers to establish connections between the video content and plaintiff Li, and it can be determined that listeners within a certain range can establish one-to-one correspondence between the disputed AI-synthesized voice and the plaintiff himself. |
| 因此，涉案声音落入李某某声音权益的保护范围。 | Therefore, the disputed voice falls within the protection scope of Li's voice rights. |
| 涉案推介视频大幅使用原告肖像、合成模拟原告的声音，未取得原告授权，故涉案视频的发布行为构成对原告肖像权和声音权益的侵犯。 | The disputed promotional video extensively used the plaintiff's likeness and synthesized voice simulating the plaintiff without obtaining the plaintiff's authorization, so the publication of the disputed video constitutes infringement of the plaintiff's portrait rights and voice rights. |
| 被告某文化传媒有限公司与视频发布者（某带货主播）依照平台规则和服务相关约定达成委托推广关系，共同为了推介被告书籍的目的发布涉案视频、获取相应收益，且被告基于平台规则和管理权限，具备对涉案视频进行审核管理的能力。 | Defendant, the cultural media company, reached a commissioning promotion relationship with the video publisher (live streaming host) according to platform rules and service-related agreements, jointly publishing the disputed video for the purpose of promoting the defendant's books and obtaining corresponding benefits. The defendant, based on platform rules and management authority, has the capability to review and manage the disputed video. |
| 在涉案视频大幅使用原告肖像、合成模拟原告声音的情况下，被告应对视频可能引发的侵权风险具有一定的预见性，并就视频是否获得了原告授权尽到合理的审查义务，但在案证据表明被告未尽到其应有审查注意义务，故被告某文化传媒有限公司应与视频发布者就侵权视频发布承担连带责任。 | In the situation where the disputed video extensively used the plaintiff's likeness and synthesized voice simulating the plaintiff, the defendant should have certain foreseeability of the potential infringement risks the video might cause and should fulfill reasonable review obligations regarding whether the video obtained the plaintiff's authorization. However, evidence in the case shows the defendant failed to fulfill its due review and attention obligations, so defendant, the cultural media company, should bear joint liability with the video publisher for the publication of the infringing video. |
| 判决被告某文化传媒有限公司向原告李某某赔礼道歉、赔偿经济损失及维权合理支出，驳回原告李某某其他诉讼请求。宣判后，双方均未提起上诉，该判决已生效。 | The court ruled that defendant, the cultural media company, should apologize to plaintiff Li, compensate for economic losses and reasonable expenses for rights protection, and dismissed plaintiff Li's other litigation claims. After the judgment, neither party appealed, and the judgment has taken effect. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 随着生成式人工智能技术的迅猛发展，名人声音被"克隆"冒用越来越真假难辨，造成声音侵权现象泛滥，消费者极易被误导。 | With the rapid development of generative artificial intelligence technology, celebrity voices being "cloned" and misused are becoming increasingly difficult to distinguish from real ones, causing rampant voice infringement phenomena and making consumers extremely susceptible to being misled. |
| 本案明确，利用AI深度合成技术合成的自然人声音，只要能够使一般社会公众或相关领域公众根据其音色、语调、发音风格等识别出特定自然人，即具有可识别性，应纳入该自然人声音权益的保护范围。 | This case clarifies that natural person voices synthesized using AI deep synthesis technology, as long as they enable the general public or relevant field public to identify specific natural persons based on timbre, intonation, pronunciation style, etc., have recognizability and should be included in the protection scope of that natural person's voice rights. |
| 同时，在商家委托视频发布者带货的法律关系中，作为委托方和实际获益者的平台商家，对其关联达人发布的推广内容负有合理审查义务，商家不能仅以"被动合作""未参与制作"为由免责，未尽到审核注意义务的，需与带货达人承担连带责任，进而为规范电商推广行为、压实商家主体责任、治理AI"声替"乱象提供规范指引，推动人工智能与深度合成技术向上向善发展。 | At the same time, in the legal relationship where merchants commission video publishers for live streaming sales, platform merchants as commissioning parties and actual beneficiaries bear reasonable review obligations for promotional content published by their affiliated influencers. Merchants cannot be exempted from liability merely on grounds of "passive cooperation" or "non-participation in production." Those failing to fulfill review and attention obligations must bear joint liability with live streaming influencers, thereby providing regulatory guidance for standardizing e-commerce promotion behavior, strengthening merchant principal responsibility, and governing AI "voice substitution" chaos, promoting the positive and virtuous development of artificial intelligence and deep synthesis technologies. |
| Chinese Original | English Translation |
|---|---|
| 案例四 | Case Four |
| 廖某诉某科技文化有限公司网络侵权责任纠纷案——未经授权对包含他人肖像的视频进行"AI换脸"处理构成对他人个人信息权益的侵害 | Liao v. A Technology Culture Co., Ltd. Network Tort Liability Dispute - Unauthorized "AI Face-Swapping" Processing of Videos Containing Others' Portraits Constitutes Infringement of Others' Personal Information Rights |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 原告廖某是一名古风短视频博主，在全网拥有较多粉丝，被告某科技文化有限公司在未经其授权同意的情况下，使用原告出镜的系列视频制作换脸模板，并上传至涉案软件中，提供给用户付费以此牟利。 | Plaintiff Liao is an ancient-style short video blogger with many followers across the internet. Defendant, a technology culture company, without plaintiff's authorization and consent, used the plaintiff's on-camera series videos to create face-swapping templates and uploaded them to the disputed software, providing them to users for payment to profit thereby. |
| 原告认为被告的行为侵犯其肖像权与个人信息权益，要求被告书面赔礼道歉、并赔偿原告经济损失与精神损失。 | The plaintiff believed the defendant's behavior infringed upon his portrait rights and personal information rights, demanding the defendant provide written apology and compensate the plaintiff for economic and mental damages. |
| 被告某科技文化有限公司认为，被告平台发布的视频均有合法来源，并且面部特征并非原告，并未侵害原告肖像权。此外，涉案软件所使用的"换脸技术"实际由第三方提供，被告并未处理原告的个人信息，并未侵害原告的个人信息权益。 | Defendant, the technology culture company, believed that the videos published on the defendant's platform all had legitimate sources, and the facial features were not the plaintiff's, so it did not infringe the plaintiff's portrait rights. Additionally, the "face-swapping technology" used by the disputed software was actually provided by a third party, the defendant did not process the plaintiff's personal information, and did not infringe the plaintiff's personal information rights. |
| 经法院查明，涉案换脸模板视频与原告创作的系列视频的妆容、发型、服饰、动作、灯光及镜头切换呈现一致特征，但出镜人的面部特征均不相同且并非原告。涉案软件通过第三方公司的服务实现换脸功能，用户交纳会员费可以解锁所有换脸功能。 | Through court investigation, it was found that the disputed face-swapping template videos showed consistent characteristics with the plaintiff's created series videos in makeup, hairstyle, clothing, movements, lighting, and camera switching, but the facial features of the people on camera were all different and not the plaintiff's. The disputed software achieved face-swapping functionality through third-party company services, and users could unlock all face-swapping functions by paying membership fees. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，判断是否侵犯肖像权的关键在于是否具备可识别性，可识别性强调肖像的本质在于指向特定的人，而肖像的范围以面部为核心，也可能涉及独特的身体部位、声音、识别性较高的特定动作等能够与特定自然人对应的部分。 | The court held that the key to determining whether portrait rights are infringed lies in whether there is recognizability. Recognizability emphasizes that the essence of a portrait lies in pointing to a specific person, and the scope of portraiture centers on the face, but may also involve unique body parts, voice, highly recognizable specific movements, and other parts that can correspond to specific natural persons. |
| 本案中，被告虽然使用原告的视频制作视频模板，但并未利用原告的肖像，而是通过技术手段将原告面部特征替换，模板中所保留的妆容、发型、服饰、灯光、镜头切换等要素并非与特定自然人不可分割，与自然人与生俱来的人格要素存在区别，一般公众通过替换后的视频识别的主体为案外人而非本案原告。 | In this case, although the defendant used the plaintiff's videos to create video templates, it did not use the plaintiff's portrait but replaced the plaintiff's facial features through technical means. The elements retained in the templates such as makeup, hairstyle, clothing, lighting, and camera switching are not inseparable from specific natural persons and are different from the inherent personality elements of natural persons. The general public would identify the subject through the replaced videos as third parties rather than the plaintiff in this case. |
| 同时，被告将视频模板提供给用户使用的行为并未丑化、污损、伪造原告肖像。因此，被告的行为不构成对原告肖像权的侵害。 | At the same time, the defendant's act of providing video templates to users did not uglify, damage, or forge the plaintiff's portrait. Therefore, the defendant's behavior does not constitute infringement of the plaintiff's portrait rights. |
| 但是，被告收集了包含原告人脸信息的出镜视频，将该视频中的原告面部替换成自己提供的照片中面部，该合成过程需要将新的静态图片中的特征与原视频部分面部特征、表情等通过算法进行融合。 | However, the defendant collected on-camera videos containing the plaintiff's facial information and replaced the plaintiff's face in those videos with faces from photos it provided. This synthesis process required fusing features from new static images with partial facial features and expressions from the original videos through algorithms. |
| 上述过程，涉及对原告个人信息的收集、使用、分析等，属于对原告个人信息的处理。被告处理该信息未经过原告同意，因此构成对原告个人信息权益的侵害。 | The above process involved collection, use, and analysis of the plaintiff's personal information, constituting processing of the plaintiff's personal information. The defendant processed this information without the plaintiff's consent, therefore constituting infringement of the plaintiff's personal information rights. |
| 对于被告擅自使用他人制作的视频侵害他人创造性劳动成果的，应由相应权利人主张权利。 | Regarding the defendant's unauthorized use of videos created by others, infringing others' creative labor achievements, the corresponding rights holders should assert their rights. |
| 判决被告向原告书面致歉，赔偿原告精神损失及精神抚慰金。宣判后，双方均未提起上诉，该判决已生效。 | The court ruled that the defendant should provide written apology to the plaintiff and compensate the plaintiff for mental damages and spiritual consolation. After the judgment, neither party appealed, and the judgment has taken effect. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 本案围绕"AI换脸"这一新商业模式，对于生成合成式人工智能应用过程中的肖像权、个人信息权益及基于劳动创造投入的合法权益进行准确区分，既维护自然人的合法权益，又为人工智能技术和新兴产业发展留有合理空间，对于服务和保障数字经济规范健康发展有重要意义。 | This case, centered around the new business model of "AI face-swapping," accurately distinguishes between portrait rights, personal information rights, and legitimate interests based on creative labor investment in the application process of generative AI, both protecting the legitimate rights of natural persons and leaving reasonable space for the development of AI technology and emerging industries, having important significance for serving and ensuring the standardized and healthy development of the digital economy. |
| Chinese Original | English Translation |
|---|---|
| 案例五 | Case Five |
| 唐某某诉某科技有限公司网络服务合同纠纷案——网络平台利用算法工具检测AI生成内容但未尽到合理适度说明义务的应承担违约责任 | Tang v. A Technology Co., Ltd. Network Service Contract Dispute - Network Platforms Using Algorithmic Tools to Detect AI-Generated Content but Failing to Fulfill Reasonable and Appropriate Explanation Obligations Should Bear Breach of Contract Liability |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 原告唐某某在被告某科技有限公司运营的网络平台上发布二百多字文字内容"打工并不能让你真的赚到多少钱，但可以开启你的新视角……如果你对学车感兴趣的话并且以后打算开车的话，可以在你最清闲的一个假期完成它……工作之后就没有太多完整的时间学驾照了"。 | Plaintiff Tang published a text content of over 200 characters on the network platform operated by defendant, a technology company: "Working doesn't really allow you to earn much money, but it can open up new perspectives for you... If you're interested in learning to drive and plan to drive in the future, you can complete it during your most leisurely vacation... After starting work, there won't be much complete time to get a driver's license." |
| 被告某科技有限公司运营的平台将该内容判定为"包含AI生成内容但未标识"的违规情况，并将该内容隐藏并作出将用户禁言一天的处理。后原告唐某某进行申诉未成功。 | The platform operated by defendant, the technology company, determined this content as a violation of "containing AI-generated content but not labeled," hid the content, and imposed a one-day ban on the user. Later, plaintiff Tang's appeal was unsuccessful. |
| 原告唐某某主张其并未使用AI创作，被告某科技有限公司的行为违约，请求法院判令被告撤销隐藏涉案内容和禁言账号一天的违规处理，并在后台系统中删除违规处理记录。 | Plaintiff Tang claimed he did not use AI for creation and that defendant's behavior constituted breach of contract, requesting the court to order the defendant to revoke the violation processing of hiding the disputed content and banning the account for one day, and delete the violation processing records in the backend system. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，网络用户在使用AI工具进行创作并将相关内容发布到网络平台时，应依照诚信原则如实标记。 | The court held that when network users use AI tools for creation and publish related content on network platforms, they should truthfully mark it according to the principle of good faith. |
| 被告发布社区公告要求创作者在发布包含AIGC生成的内容时，应主动使用标识，对发布时未主动声明的内容，平台将采取适当措施进行流通限制并添加相关标识；未经标签声明，使用人工智能生成的内容构成违规。 | The defendant published community announcements requiring creators to actively use labels when publishing content containing AIGC generation. For content not actively declared during publication, the platform would take appropriate measures to restrict circulation and add relevant labels; using AI-generated content without label declaration constitutes a violation. |
| 原告系平台注册用户，上述公告属于平台服务协议的组成部分，被告有权根据协议对用户发布内容是否属于AI生成合成内容进行审查和处理。 | The plaintiff is a registered user of the platform, and the above announcement is part of the platform service agreement. The defendant has the right to review and process whether user-published content belongs to AI-generated synthetic content according to the agreement. |
| 在一般情况下，原告应该提供如创作底稿、原件、源文件、源数据等证明人类创作属性的初步证据，但本案中原告回答内容系即时性创作文本，客观上无法提供上述证据，由其提供上述证据不存在合理性和现实可行性。 | In general circumstances, the plaintiff should provide preliminary evidence proving human creative attributes such as creative drafts, originals, source files, source data, etc. However, in this case, the plaintiff's response content was instantaneous creative text, objectively unable to provide the above evidence, and providing such evidence lacks reasonableness and practical feasibility. |
| 被告基于算法工具的运算结果得出涉案内容为AI生成合成的审查结论，被告既是算法工具掌控方又是结果判断方，算法工具运算过程和审查结果由被告掌控，被告应就该事实进行合理举证或解释说明。 | The defendant reached the review conclusion that the disputed content was AI-generated and synthesized based on algorithmic tool calculation results. The defendant is both the controller of the algorithmic tool and the judge of results, with the algorithmic tool's calculation process and review results controlled by the defendant. The defendant should provide reasonable evidence or explanatory clarification regarding this fact. |
| 虽然被告提供了算法的备案信息，但无法确认备案信息与本案争议的关联性。被告未对算法决策依据和结果进行适度的解释和说明，未对涉案内容属于AI生成合成的判断依据和结果作出合理解释，应对其没有事实依据的情况下对涉案账户的处理行为承担违约责任。 | Although the defendant provided filing information for the algorithm, the relevance between the filing information and this case's dispute cannot be confirmed. The defendant failed to provide appropriate explanation and clarification of the algorithmic decision basis and results, failed to provide reasonable explanation for the judgment basis and results that the disputed content belongs to AI generation and synthesis, and should bear breach of contract liability for processing the disputed account without factual basis. |
| 被告进行人工复核的标准要求具有非常明显的人类情感特征，该判断并不具有科学依据以及较强的说服力和可信度。 | The defendant's manual review standard requires very obvious human emotional characteristics, which lacks scientific basis and strong persuasiveness and credibility. |
| 法院判决被告展开折叠的回答并删除相关后台记录。被告不服一审判决提起上诉，后撤回上诉，一审判决生效。 | The court ruled that the defendant should unfold the collapsed answer and delete relevant backend records. The defendant appealed the first-instance judgment but later withdrew the appeal, and the first-instance judgment took effect. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 本案是对AI生成合成内容标识的标记、平台识别和治理在司法审查规则方面的有益探索，一方面肯定了网络内容服务平台利用算法工具对AI生成合成内容进行审查和处理，履行信息内容管理主体责任的积极作用； | This case is a beneficial exploration of judicial review rules for marking, platform identification, and governance of AI-generated synthetic content labels. On one hand, it affirms the positive role of network content service platforms using algorithmic tools to review and process AI-generated synthetic content, fulfilling their principal responsibilities for information content management; |
| 另一方面，认定网络内容服务平台在即时性创作文本场景下对使用算法自动化决策结果承担适度说明义务，并为司法审查阶段的算法解释说明程度确立了一定的标准。 | On the other hand, it determines that network content service platforms bear appropriate explanation obligations for using algorithmic automated decision results in instantaneous creative text scenarios, and establishes certain standards for the degree of algorithmic explanation and clarification in the judicial review stage. |
| 本案通过司法个案裁判，合理分配网络内容服务平台和用户的举证责任，推动网络内容服务平台提高算法的识别能力和决策能力，切实提升人工智能信息内容治理水平。 | This case, through individual judicial case judgment, reasonably allocates the burden of proof between network content service platforms and users, promotes network content service platforms to improve algorithmic recognition and decision-making capabilities, and effectively enhances the governance level of AI information content. |
| Chinese Original | English Translation |
|---|---|
| 案例六 | Case Six |
| 程某诉孙某网络侵权责任纠纷案——利用AI软件恶搞、丑化他人肖像构成人格权侵权 | Cheng v. Sun Network Tort Liability Dispute - Using AI Software to Spoof and Uglify Others' Portraits Constitutes Personality Rights Infringement |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 原告程某与被告孙某同为摄影交流微信群成员，被告孙某未经原告程某同意，使用AI软件将原告程某用作微信头像的肖像照片生成衣着暴露的动漫风格图片，并发送至该摄影交流微信群内，后原告程某多次制止，被告孙某仍继续使用涉案AI软件将原告程某的上述微信头像肖像照片AI生成衣着暴露且身体畸形的动漫风格图片并微信私信原告。 | Plaintiff Cheng and defendant Sun were both members of a photography exchange WeChat group. Without plaintiff Cheng's consent, defendant Sun used AI software to generate anime-style images with revealing clothing from plaintiff Cheng's portrait photo used as WeChat avatar and sent them to the photography exchange WeChat group. After plaintiff Cheng's multiple attempts to stop this, defendant Sun continued to use the disputed AI software to generate anime-style images with revealing clothing and deformed bodies from plaintiff Cheng's WeChat avatar portrait photos and privately messaged the plaintiff via WeChat. |
| 原告程某认为被告孙某群发和私信的上述被诉侵权图片，能够识别出为原告本人形象，且图片带有严重的性暗示和丑化性质，使得原告的社会评价降低，构成对原告肖像权、名誉权及一般人格权的侵害。 | Plaintiff Cheng believed that the above-mentioned disputed infringing images sent by defendant Sun in group messages and private messages could be identified as the plaintiff's own image, and the images carried serious sexual implications and uglifying nature, causing the plaintiff's social evaluation to decline, constituting infringement of the plaintiff's portrait rights, reputation rights, and general personality rights. |
| 原告程某据此要求被告赔礼道歉，并赔偿精神损失和经济损失。 | Based on this, plaintiff Cheng demanded the defendant apologize and compensate for mental and economic damages. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，被告孙某发布于微信群内的被诉侵权图片系擅自利用原告程某微信头像肖像照片AI生成，与原告程某微信头像肖像照片所呈现的自然人外部形象在脸型、姿势、造型等方面均具有高度的对应关系，微信群成员能够通过上述人物体貌及群聊的前后语境识别出被诉侵权图片的人物外部形象所对应的主体为原告，因此被告的群发行为构成对原告肖像权的侵害。 | The court held that the disputed infringing images published by defendant Sun in the WeChat group were AI-generated by unauthorized use of plaintiff Cheng's WeChat avatar portrait photos, having high correspondence with the external image of the natural person presented in plaintiff Cheng's WeChat avatar portrait photos in terms of face shape, posture, styling, etc. WeChat group members could identify through the above character's physical appearance and the context of group chat that the subject corresponding to the external image of the character in the disputed infringing images was the plaintiff. Therefore, the defendant's group messaging behavior constitutes infringement of the plaintiff's portrait rights. |
| 原告通过微信头像展示的个人肖像照片是其网络虚拟身份的标识。被告利用AI软件生成的上述被诉侵权图片将原告穿着得体的微信头像肖像照片生成为胸部暴露的图片，引发了微信群内针对原告的不当讨论，客观上导致了他人对原告的低俗化评价，构成对原告肖像权、名誉权的侵害。 | The personal portrait photos displayed through the plaintiff's WeChat avatar are identifiers of his virtual network identity. The above-mentioned disputed infringing images generated by the defendant using AI software transformed the plaintiff's decently dressed WeChat avatar portrait photos into images with exposed chest, triggering inappropriate discussions about the plaintiff within the WeChat group, objectively leading to vulgarized evaluations of the plaintiff by others, constituting infringement of the plaintiff's portrait rights and reputation rights. |
| 此外，被告孙某利用AI软件将原告程某微信头像肖像照片生成人物腿部为木头状，甚至存在三只手臂的图片，图片中的人物体型明显不符合人类基本身体结构，亦将人物的胸部进行了暴露展示。 | In addition, defendant Sun used AI software to generate images from plaintiff Cheng's WeChat avatar portrait photos where the character's legs were wooden-like and even had three arms. The character's body type in the images obviously did not conform to basic human body structure, and also exposed the character's chest. |
| 被告将上述图片私信原告，势必会造成原告心理屈辱，侵害了原告的人格尊严，构成对原告一般人格权的侵害。 | The defendant's private messaging of the above images to the plaintiff would inevitably cause psychological humiliation to the plaintiff, infringing the plaintiff's personal dignity and constituting infringement of the plaintiff's general personality rights. |
| 判决孙某向程某公开赔礼道歉，赔偿原告精神损失。宣判后，双方均未提起上诉，该判决已生效。 | The court ruled that Sun should publicly apologize to Cheng and compensate the plaintiff for mental damages. After the judgment, neither party appealed, and the judgment has taken effect. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 本案中，法院认定行为人未经许可利用AI软件恶搞、丑化他人肖像构成对他人人格权的侵害，强调生成式人工智能技术的使用者应当遵守法律、行政法规，尊重社会公德和伦理道德，尊重他人合法权益，不得危害他人身心健康，明确了普通网络用户使用生成式人工智能技术的行为边界，对人工智能时代下加强自然人人格权保护具有典型意义。 | In this case, the court determined that actors using AI software to spoof and uglify others' portraits without permission constitutes infringement of others' personality rights, emphasizing that users of generative artificial intelligence technology should comply with laws and administrative regulations, respect social morality and ethical standards, respect others' legitimate rights and interests, and must not harm others' physical and mental health. This clarifies the behavioral boundaries for ordinary network users using generative AI technology and has typical significance for strengthening protection of natural persons' personality rights in the AI era. |
| Chinese Original | English Translation |
|---|---|
| 案例七 | Case Seven |
| 某甲科技有限公司、某乙科技有限公司诉孙某某、某网络科技有限公司著作权权属、侵权纠纷案——具有独创性的虚拟数字人形象构成美术作品 | Company A Technology Co., Ltd. and Company B Technology Co., Ltd. v. Sun and A Network Technology Co., Ltd. Copyright Ownership and Infringement Dispute - Virtual Digital Human Images with Originality Constitute Fine Art Works |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 虚拟数字人甲、乙由原告某甲科技有限公司、原告某乙科技有限公司等四家单位联合制作，其中原告某甲科技有限公司为著作权人，原告某乙科技有限公司为被许可人。 | Virtual digital humans A and B were jointly created by four units including plaintiff Company A Technology Co., Ltd. and plaintiff Company B Technology Co., Ltd., with plaintiff Company A Technology Co., Ltd. as the copyright owner and plaintiff Company B Technology Co., Ltd. as the licensee. |
| 虚拟数字人甲在各平台拥有超440万粉丝，曾获评2022年文化产业和旅游业年度八大热点事件。 | Virtual digital human A has over 4.4 million followers across various platforms and was rated as one of the eight major hot events in the cultural industry and tourism industry in 2022. |
| 二原告主张，虚拟数字人甲、乙形象构成美术作品，虚拟数字人甲形象首次发表于短剧《千***》第一集，虚拟数字人乙形象首次发表于微博账号"之****"。 | The two plaintiffs claimed that the images of virtual digital humans A and B constitute fine art works, with virtual digital human A's image first published in the first episode of the short drama "Qian***," and virtual digital human B's image first published on the Weibo account "Zhi****." |
| 被告某联合创作单位员工孙某某离职后，在被告某网络科技有限公司运营的某模型网上擅自售卖虚拟数字人甲、乙模型，侵害了二原告就虚拟数字人形象享有的复制权和信息网络传播权，被告某网络科技有限公司作为平台方未尽到监管责任，应与被告孙某某承担连带责任。 | After defendant Sun, an employee of a joint creation unit, left his position, he unauthorized sold virtual digital human A and B models on a model website operated by defendant, a network technology company, infringing the two plaintiffs' reproduction rights and information network dissemination rights regarding the virtual digital human images. Defendant, the network technology company, as a platform, failed to fulfill supervisory responsibilities and should bear joint liability with defendant Sun. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，虚拟数字人甲全身形象和乙的头部形象，并不直接来源于真人，而是由制作团队制作，具有明显的艺术创作效果，体现了制作团队对线条、色彩和具体形象设计的独特的美学选择和判断，具备作品的独创性要求，构成美术作品。 | The court held that the full-body image of virtual digital human A and the head image of B do not directly derive from real people but were created by the production team, having obvious artistic creative effects, reflecting the production team's unique aesthetic choices and judgments regarding lines, colors, and specific image design, meeting the originality requirements of works and constituting fine art works. |
| 被告孙某某在某模型网发布被诉侵权模型，在人物五官、发型、发饰、服装的设计及整体风格，尤其是在权利作品具有独创性的元素组合方面，与涉案作品虚拟数字人甲、乙形象相同或相似，可以认定构成实质性相似，侵害了二原告对涉案作品享有的信息网络传播权。 | Defendant Sun published the disputed infringing models on a model website. In terms of character facial features, hairstyle, hair accessories, clothing design, and overall style, especially in the combination of original elements that the rights works possess, they are identical or similar to the disputed works' virtual digital human A and B images, which can be determined to constitute substantial similarity, infringing the two plaintiffs' information network dissemination rights regarding the disputed works. |
| 综合考量被告某网络科技有限公司服务的具体类型、对被诉内容的干预程度、是否直接获得经济利益、权利作品的知名度、被诉内容的热度等因素，被告某网络科技有限公司作为网络服务提供者不构成共同侵权。 | Comprehensively considering factors such as the specific type of services of defendant, the network technology company, the degree of intervention in the disputed content, whether direct economic benefits were obtained, the popularity of the rights works, and the popularity of the disputed content, defendant, the network technology company, as a network service provider, does not constitute joint infringement. |
| 虚拟数字人承载多重权益，本案仅就美术作品的权益进行认定，综合考虑请求保护的权利类型、市场价值和侵权人主观过错、侵权行为性质和规模、损害后果严重程度确定本案的经济赔偿金额。 | Virtual digital humans carry multiple rights and interests. This case only determined the rights and interests of fine art works, comprehensively considering the types of rights for which protection is requested, market value, and the infringer's subjective fault, the nature and scale of the infringement, and the severity of damage consequences to determine the economic compensation amount for this case. |
| 判决被告孙某某赔偿二原告经济损失。被告孙某某不服一审判决，提起上诉。二审法院判决驳回上诉，维持原判。 | The court ruled that defendant Sun should compensate the two plaintiffs for economic losses. Defendant Sun was dissatisfied with the first-instance judgment and appealed. The second-instance court ruled to dismiss the appeal and uphold the original judgment. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 本案涉及虚拟数字人形象的法律属性及独创性认定。虚拟数字人由外在表现及技术内核两部分组成，具有数字化外形与类人化功能。 | This case involves the legal nature and originality determination of virtual digital human images. Virtual digital humans consist of two parts: external manifestation and technical core, having digitized appearance and humanoid functions. |
| 在外在表现层面，如体现了制作团队对线条、色彩和具体形象设计的独特的美学选择和判断，具备作品的独创性要求，可以认定为美术作品，受到著作权法保护。 | At the external manifestation level, if they reflect the production team's unique aesthetic choices and judgments regarding lines, colors, and specific image design, meeting the originality requirements of works, they can be determined as fine art works and protected by copyright law. |
| 本案为同类案件裁判提供了借鉴，有利于助力虚拟数字人产业繁荣、推动新质生产力发展。 | This case provides reference for similar case judgments, helping to promote the prosperity of the virtual digital human industry and advance the development of new quality productive forces. |
| Chinese Original | English Translation |
|---|---|
| 案例八 | Case Eight |
| 何某诉某人工智能科技有限公司网络侵权责任纠纷案——未经同意创造自然人的AI形象构成对人格权的侵害 | He v. An AI Technology Co., Ltd. Network Tort Liability Dispute - Creating AI Images of Natural Persons Without Consent Constitutes Infringement of Personality Rights |
| Chinese Original | English Translation |
|---|---|
| 基本案情 | Basic Case Facts |
| 某人工智能科技有限公司（以下简称被告）是某款手机记账软件的开发运营者，用户在该软件中可自行创设"AI陪伴者"，设置陪伴者的名称、头像，设置与该陪伴者的人物关系（如男女朋友、兄妹、母子等）。 | An AI technology company (hereinafter referred to as the defendant) is the developer and operator of a mobile accounting software, in which users can create "AI companions," set the companion's name and avatar, and set character relationships with the companion (such as boyfriend/girlfriend, siblings, mother and son, etc.). |
| 何某系公众人物，知名度较高，在该款软件中被大量用户设置为陪伴人物。用户在设置"何某"为陪伴人物时，上传了大量何某的肖像图片设置人物头像，同时设置了人物关系。 | He is a public figure with high popularity and was set as a companion character by numerous users in this software. When users set "He" as a companion character, they uploaded numerous portrait images of He to set as character avatars and also set character relationships. |
| 被告通过算法部属，将陪伴者"何某"按关系设定分类，并向其他用户推荐该角色。为了使AI角色更加拟人化，被告还为AI角色提供了"调教"算法机制，即用户上传各类文字、肖像图片、动态表情等互动语料，部分用户参与审核，软件进行筛选、分类，形成人物语料。 | The defendant, through algorithm deployment, classified the companion "He" according to relationship settings and recommended this character to other users. To make AI characters more humanized, the defendant also provided a "training" algorithm mechanism for AI characters, where users upload various texts, portrait images, dynamic expressions, and other interactive materials, with some users participating in review, and the software screening and categorizing to form character materials. |
| 软件可以根据话题类别、角色的人设特点等，在"何某"与用户的对话中向用户推送与其有关的"肖像表情包"和"撩人情话"，营造出与其真人真实互动的使用体验。 | The software can push "portrait emoji packages" and "flirtatious words" related to him to users in conversations between "He" and users based on topic categories and character personality traits, creating a user experience of real interaction with the real person. |
| 原告认为，被告的行为侵害了原告的姓名权、肖像权、一般人格权，故诉至法院。 | The plaintiff believed the defendant's behavior infringed the plaintiff's name rights, portrait rights, and general personality rights, and thus filed a lawsuit in court. |
| Chinese Original | English Translation |
|---|---|
| 裁判结果 | Judgment Result |
| 法院认为，被告行为构成对原告姓名权、肖像权、一般人格权的侵害。 | The court held that the defendant's behavior constitutes infringement of the plaintiff's name rights, portrait rights, and general personality rights. |
| 在被告的软件功能和算法设计下，用户使用何某的姓名、肖像创设虚拟人物，制作互动语料素材，将何某的姓名、肖像、人格特点等综合而成的整体形象投射到AI角色上，形成了何某的虚拟形象，是对包含何某肖像、姓名的整体人格形象的使用。 | Under the defendant's software functions and algorithm design, users used He's name and portrait to create virtual characters and produce interactive material, projecting the overall image composed of He's name, portrait, personality characteristics, etc., onto AI characters, forming He's virtual image, which is use of the overall personality image containing He's portrait and name. |
| 被告使用何某姓名、肖像的行为并未获得何某的许可，构成对何某姓名权、肖像权的侵害。 | The defendant's use of He's name and portrait was not authorized by He, constituting infringement of He's name rights and portrait rights. |
| 同时，用户可以与AI角色设置虚拟身份关系，设定任意相互称谓，通过制作语料素材"调教"角色，使得AI角色与真实自然人高度关联，容易让用户产生一种与何某真实互动的情感体验。 | At the same time, users can set virtual identity relationships with AI characters, establish arbitrary mutual titles, and "train" characters through creating material, making AI characters highly associated with real natural persons, easily causing users to have emotional experiences of real interaction with He. |
| 上述使用方式未经何某同意，侵害了何某的人格尊严和人格自由利益，构成对何某一般人格权益的侵害。 | The above usage methods were not consented to by He, infringing He's personal dignity and personal freedom interests, constituting infringement of He's general personality rights and interests. |
| 同时，涉案软件的服务与技术服务存在本质不同，被告并非提供简单的用户内容上传服务，而是通过规则设定、算法设计，组织、鼓励用户形成侵权素材，与其共同创设虚拟形象，并使用到用户服务中。 | At the same time, the disputed software's services are essentially different from technical services. The defendant did not provide simple user content upload services but organized and encouraged users to form infringing materials through rule setting and algorithm design, jointly creating virtual images with them and using them in user services. |
| 被告的产品设计和算法应用鼓励、组织了涉案虚拟形象的创设，直接决定了软件核心功能的实现，不再是中立的技术服务提供者，而应作为网络内容服务提供者承担侵权责任。 | The defendant's product design and algorithm application encouraged and organized the creation of the disputed virtual images, directly determining the realization of the software's core functions. It is no longer a neutral technical service provider but should bear infringement liability as a network content service provider. |
| 判决被告向原告公开赔礼道歉，赔偿原告精神损失及经济损失。被告不服一审判决提起上诉，后撤回上诉，一审判决生效。 | The court ruled that the defendant should publicly apologize to the plaintiff and compensate the plaintiff for mental and economic damages. The defendant appealed the first-instance judgment but later withdrew the appeal, and the first-instance judgment took effect. |
| Chinese Original | English Translation |
|---|---|
| 典型意义 | Typical Significance |
| 本案明确自然人的人格权及于其虚拟形象，未经许可擅自创设、使用自然人虚拟形象的，构成对自然人人格权的侵害； | This case clarifies that natural persons' personality rights extend to their virtual images, and unauthorized creation and use of natural persons' virtual images constitutes infringement of natural persons' personality rights; |
| 网络服务提供者通过算法设计实质参与了侵权内容的生成和提供，应按照内容服务提供者承担侵权责任。 | Network service providers that substantially participate in the generation and provision of infringing content through algorithm design should bear infringement liability as content service providers. |
| 本案对加强人格权保护具有重要意义，入选最高人民法院"民法典颁布后人格权司法保护典型民事案例"、北京法院参阅案例。 | This case has important significance for strengthening personality rights protection and was selected as a "Typical Civil Case for Judicial Protection of Personality Rights After the Promulgation of the Civil Code" by the Supreme People's Court and as a reference case for Beijing courts. |