声音和形象被AI合成滥用的事儿正变得越来越多。比如本周，奥运冠军全红婵、王楚钦等人的声音被 AI克隆，卖起了所谓的"土鸡蛋"。不少配音演员也站出来维权，因为他们赖以生存的声音，如今可以被 AI一键复制，随意地使用。 这不禁让我们越来越担心，当我们的声音和面容到了如此轻易被人挪用的地步，人与人之间的信任该如 何维系？AI的发展速度正越来越快，我们该如何用监管和立法守住底线？《新闻周刊》本周关注：人 工智能的滥用，到底怎么治？ AI换脸、声音克隆等合成技术滥用 让"耳听眼见"未必为实 这是配音演员孙晨铭的原声，而这段，是某自媒体账号通过AI克隆其声音生成的配音作品。如果没有 提示，谁能分辨出两者之间的差异呢？也是在最近，孙晨铭还发现，一家房产公司同样未经授权，就对 他的配音作品进行了AI克隆，用在了公司的企业宣传片中。 声音被AI克隆，并被用于各种各样的商业场景，对于孙晨铭和同行们来说，已经不是什么新鲜事儿。 有人给孙晨铭发来民法典的法条，声音权和肖像权一样，是公民不可侵犯的权益。于是，不少配音演员 自发到那条企业宣传片的评论区集体留言，压力之下，这家房产公司，向孙晨铭支付了版权费用。 本周，本台也曝光了利用AI ...

近日，有电商平台商家在社交媒体上反映，个别消费者利用AI工具生成虚假的商品瑕疵图片，要求商家"仅退款"。 "还好截屏快，AI的水印还没去掉就水灵灵地发过来了。"电商客服可可（化名）在社交平台分享了自己的工作遭遇。 2025年6月25日，网友在社交媒体分享收到的售后图片。（图片来源：网络） 这类情况并非个例。许多商家也遇到类似问题，消费者提供的视频上突然出现两只"右手"；水果上出现明显伪造的霉斑…… 张毅说："这种行为增加了商家的审核和经营成本，许多商家只好选择直接退货或赔付，而举证维权的难度也随之加大。长 期来看，这将影响平台的管理成本、运营成本以及信用机制。对电商行业的健康有序发展构成挑战。" 如何解决？ 今年以来，国家层面已就AI滥用现象出台了相关办法。3月，四部门联合发布《人工智能生成合成内容标识办法》（以下简 称《标识办法》）。《标识办法》强调，任何组织和个人不得恶意删除、篡改、伪造、隐匿本办法规定的生成合成内容标识，不 得为他人实施上述恶意行为提供工具或者服务，不得通过不正当标识手段损害他人合法权益。《标识办法》自2025年9月1日起施 行。4月，中央网信办在全国范围内部署开展为期3个月的"清朗·整 ...

Core Viewpoint - The Shanghai Cyberspace Administration has initiated penalties against several generative artificial intelligence service websites for failing to comply with legal safety assessment requirements and for not implementing necessary safety measures to prevent the generation of illegal content [1] Group 1: Regulatory Actions - The Shanghai Cyberspace Administration discovered that certain websites providing generative AI services did not conduct safety assessments as mandated by law [1] - These websites failed to take necessary precautions to prevent the generation of illegal content, including violations of personal information rights and the production of illicit materials such as money laundering content and pornographic images [1] - Companies are urged to take down related functionalities and may only resume operations after passing safety evaluations [1] Group 2: Future Enforcement Focus - The Shanghai Cyberspace Administration will continue to combat the misuse of AI, particularly focusing on issues related to "AI disguise," "AI face-swapping and voice-changing," and "AI forgery" [1] - There will be a concentrated effort to address algorithmic services that infringe on personal information rights, with strict penalties for repeat offenders and those with serious issues [1]

Group 1: Kimi's Developments - Kimi announced a copyright collaboration with Caixin Media, becoming the first domestic company to publicly declare such a partnership with a media outlet [1] - Kimi has launched a quality source badge system to help users identify high-quality information sources, focusing on government and educational websites [1][2] - The collaboration will allow Kimi to generate answers based on Caixin's professional reporting when users inquire about financial topics [1] Group 2: Anthropic's Warning - Anthropic reported that its AI model Claude is being misused for "influence-as-a-service," with organizations creating over 100 bot accounts to spread politically biased content [3][4] - The misuse includes not only content generation but also automated interactions such as commenting and sharing, indicating a sophisticated manipulation of social media [4] Group 3: Adobe's New Application - Adobe launched a new application called Content Authenticity, allowing users to protect images from AI training by embedding invisible metadata and a "prohibit AI training" label [5] - The application is currently in public testing and supports batch processing of up to 50 images, but Adobe has not yet signed agreements with AI model creators to enforce this standard [5][6] Group 4: Weibo's AI Privacy Response - Weibo's AI feature "AI Smart Search" faced backlash for allegedly infringing on user privacy, prompting the company to clarify that it does not analyze non-public user data [7] - Following user feedback, Weibo has adjusted the functionality of the AI Smart Search to avoid presenting content that may cause discomfort to users [7] Group 5: E-commerce Platforms' Policy Change - Major e-commerce platforms, including Pinduoduo, Taobao, Douyin, Kuaishou, and JD.com, are set to eliminate the "refund only" policy, allowing merchants to handle refund requests independently [8] - These platforms have engaged in discussions with regulatory authorities regarding the cancellation of the "refund only" policy and will announce the details publicly once finalized [8]