YouTube is expanding its AI deepfake detection tool to all adult users
YouTube is expanding its AI deepfake detection tool to all adult users
YouTube 将其 AI 深度伪造检测工具扩展至所有成年用户
YouTube is expanding its AI likeness detection program to all users over the age of 18 — meaning just about anyone can have the platform hunt for potential deepfakes of themselves. YouTube 正在将其 AI 肖像检测程序扩展至所有 18 岁以上的用户，这意味着几乎任何人都可以让该平台协助搜寻可能存在的个人深度伪造内容。
The likeness detection feature uses a selfie-style scan of a person’s face to monitor YouTube for lookalikes. If there is a match, YouTube alerts the user; the person then has the option to request that YouTube remove the content. YouTube has said in the past that it has found the number of removal requests to be “very small.” 该肖像检测功能通过类似自拍的扫描方式获取用户面部信息，从而在 YouTube 上监测是否存在相似面孔。如果发现匹配内容，YouTube 会向用户发出提醒；随后，用户可以选择要求 YouTube 删除该内容。YouTube 此前曾表示，目前收到的删除请求数量“非常少”。
YouTube began testing the feature with content creators, and then expanded it to government officials, politicians, journalists, and finally the entertainment industry. The expansion to any user 18 years or older is a significant shift — it essentially gives the average person the ability to constantly monitor content on YouTube that could use their likeness. YouTube 最初在内容创作者中测试该功能，随后扩展至政府官员、政治人物、记者，最后覆盖到娱乐行业。此次向所有 18 岁及以上用户开放是一个重大转变——它实质上赋予了普通人持续监测 YouTube 上可能使用其肖像的内容的能力。
Takedown requests are evaluated using YouTube’s privacy policy, and the company says it considers removals based on criteria like whether the content is realistic, is labeled as AI-generated, and if a person can be uniquely identified. There are carveouts for things like parody or satire, and the tool only covers facial likeness, not other identifying features like a person’s voice. Users can withdraw from the program and have YouTube delete their data. 下架请求将根据 YouTube 的隐私政策进行评估。公司表示，审核删除请求时会考虑多项标准，例如内容是否逼真、是否标注为 AI 生成，以及是否能唯一识别出特定个人。该政策对戏仿或讽刺内容设有豁免条款，且该工具仅涵盖面部肖像，不包括声音等其他识别特征。用户可以随时退出该计划，并要求 YouTube 删除其相关数据。
The news was announced on YouTube’s creator forum, but spokesperson Jack Malon says there are no requirements on what constitutes a “creator” who is eligible. 这一消息是在 YouTube 的创作者论坛上宣布的，但发言人 Jack Malon 表示，对于何种身份才算符合条件的“创作者”，并没有具体要求。
“With this expansion, we’re making clear that whether creators have been uploading to YouTube for a decade or are just starting, they’ll have access to the same level of protection,” Malon said in an email. Malon 在一封电子邮件中表示：“通过此次扩展，我们明确了无论创作者是在 YouTube 上上传了十年的老用户，还是刚刚起步的新人，他们都将获得同等水平的保护。”
Deepfake content often centers on celebrities, politicians, or other public figures, but the ability to create a convincing digital replica is a concern for private citizens, too. There have been instances of teenagers being deepfaked by classmates, and three teenagers sued xAI alleging that the company’s Grok chatbot generated child sexual abuse material (CSAM) of them. 深度伪造内容往往集中在名人、政治人物或其他公众人物身上，但创建逼真数字副本的能力也让普通公民感到担忧。此前曾发生过青少年被同学深度伪造的案例，还有三名青少年起诉 xAI，指控该公司的 Grok 聊天机器人生成了关于他们的儿童性虐待材料（CSAM）。