Meta ran ads for an app promising to nudify female politicians
Meta ran ads for an app promising to nudify female politicians
Meta 投放广告推广一款承诺将女性政客“脱衣化”的应用程序
Meta platforms recently ran ads for an AI porn-generation tool that seemingly encouraged users to create deepfaked videos resembling female US politicians, despite the company’s policies against ads containing sexual material. It’s the latest in a series of failures by Meta to keep advertisements for tools that produce nonconsensual intimate imagery off its platforms.
Meta 平台近期投放了针对一款 AI 色情生成工具的广告，该工具似乎鼓励用户制作酷似美国女性政客的深度伪造（deepfake）视频，尽管该公司明文禁止包含性内容的广告。这是 Meta 在阻止推广制作非自愿私密影像工具的广告方面，一系列失败案例中的最新一起。
The tool, which is called Kromix, bills itself as an “AI image styler.” A voice-over for one of the video advertisements, viewed by WIRED, promotes the tool as having “no restrictions,” saying “all the characters are real people” and describing it as “the AI that men actually use.”
这款名为 Kromix 的工具自称为“AI 图像风格化器”。据《连线》（WIRED）查看的一则视频广告配音显示，该工具宣称“没有任何限制”，并称“所有角色都是真实存在的人”，将其描述为“男性真正使用的 AI”。
In one ad, a woman closely resembling a prominent female US politician is seen in front of a US flag and the flag of the president of the US above the caption “What if she moved?” She then winks, before the ad cuts to a scene of a woman with the same face performing in a pornographic video.
在一则广告中，一名酷似美国知名女性政客的女子出现在美国国旗和美国总统旗帜前，上方配文写着“如果她动起来会怎样？”随后她眨了眨眼，画面随即切换到一名拥有相同面孔的女性在色情视频中表演的场景。
The Kromix app, which was hosted by Apple’s App Store prior to a request for comment from WIRED, invites paid users to upload photos so that it can use them in scenarios including “bedroom rape” and “Disney love.” Videos on the app also feature AI-generated pornographic videos of women wearing Spider-Man costumes. Requests for comment sent to a contact listed in the app went unanswered.
在《连线》发出置评请求之前，Kromix 应用程序曾托管在苹果 App Store 中。该应用诱导付费用户上传照片，以便将其用于包括“卧室强奸”和“迪士尼之恋”等场景中。该应用内的视频还包含女性身穿蜘蛛侠服装的 AI 生成色情内容。发送至应用内所列联系人的置评请求未获回应。
Apple has struggled to keep similar tools, which violate its policies against nonconsensual intimate imagery, nudifying, and sexually explicit content, off the App Store. San Francisco’s city attorney sent the company a cease-and-desist letter earlier this summer demanding the removal of 13 specific apps and that it stop “aiding and abetting” the sale of explicit deepfake images.
苹果公司在将此类工具拒之门外方面一直面临困难，这些工具违反了其关于非自愿私密影像、“脱衣化”及性露骨内容的政策。今年夏天早些时候，旧金山市检察官向苹果公司发送了一封停止侵权函，要求其下架 13 款特定应用程序，并停止“协助和教唆”销售露骨的深度伪造图像。
The ads were, according to Meta’s ad library, exclusively targeted to male users with the tagline “Unleash Next-gen AI Magic” and were removed after WIRED asked about them. In all, the account tied to the campaign ran 32 ads, according to Meta, which ran for between five and 46 hours. Most received 10 or fewer impressions.
根据 Meta 的广告资料库显示，这些广告专门针对男性用户投放，标语为“释放下一代 AI 魔法”，在《连线》询问后被删除。据 Meta 称，与该活动关联的账户总共投放了 32 则广告，运行时间在 5 到 46 小时之间。大多数广告的展示次数在 10 次或以下。
Katie Paul, director of the Tech Transparency Project, which discovered the ads, says, “All of the ads that we’ve seen so far broadly on the nudify issue are exclusively sexualizing and nudifying and putting women into AI-generated porn. And this raises a lot of questions about how Meta is going to deal with this kind of content, not just broadly but when it comes to female politicians, when the company is actively running ads containing this content on their platform.”
发现这些广告的“技术透明度项目”（Tech Transparency Project）总监凯蒂·保罗（Katie Paul）表示：“到目前为止，我们所看到的关于‘脱衣化’问题的广告，全部都是为了性化、脱衣化，并将女性置于 AI 生成的色情内容中。这引发了许多关于 Meta 将如何处理此类内容的质疑——不仅是广泛意义上的处理，更是在涉及女性政客时，当该公司在其平台上积极投放包含此类内容的广告时，该如何应对。”
Politicians and elected officials, overwhelmingly women, have been a target of nonconsensual deepfake images and videos for years. In January, as Grok was used to create around 3 million sexualized images of women and girls, officials in Sweden, England, and Northern Ireland were targeted by people using the technology, often using official images as a starting point.
多年来，政客和民选官员（绝大多数为女性）一直是未经同意的深度伪造图像和视频的目标。今年 1 月，当 Grok 被用于生成约 300 万张女性和女孩的性化图像时，瑞典、英国和北爱尔兰的官员也成为了该技术的使用目标，攻击者通常以官方照片作为起点。
“As the production and distribution of AI-generated content have democratized, formerly benign media—official portraits, video from a press conference—have become raw material for low- and no-code tools that anyone with an internet connection can use,” says Benjamin Shultz, the lead researcher at the American Sunlight Project, who is investigating how the ecosystem of harmful deepfake technology is used to target politicians.
“随着 AI 生成内容的生产和分发变得平民化，曾经无害的媒体素材——如官方肖像、新闻发布会视频——已成为低代码和无代码工具的原材料，任何能上网的人都可以使用，”美国阳光项目（American Sunlight Project）首席研究员本杰明·舒尔茨（Benjamin Shultz）说道。他目前正在调查有害深度伪造技术生态系统如何被用于针对政客。
According to the company’s public documentation, before an ad is published on Meta’s platforms, which includes Instagram and Facebook, each ad is “reviewed” against its policies that detail what advertisers can and can’t post. “Our ad review system relies primarily on automated tools to check ads and business assets against our policies,” the company’s advertising standards say, pointing out that the review starts before ads begin running.
根据该公司的公开文档，在广告发布到 Meta 平台（包括 Instagram 和 Facebook）之前，每则广告都会根据其详细规定广告主可发布和不可发布内容的政策进行“审核”。该公司的广告标准称：“我们的广告审核系统主要依赖自动化工具，根据我们的政策检查广告和商业资产。”并指出审核在广告投放前即已开始。
The page running the advertisements was created on August 3, not long before the ads began running, and had no followers—two things that might have indicated to Meta that the account was suspicious. The company specifically has a policy that says ads cannot contain sexual imagery—including nudity, sexual activity, or sexually suggestive content.
投放这些广告的页面创建于 8 月 3 日，即广告开始投放前不久，且没有粉丝——这两点本应向 Meta 表明该账户存在可疑之处。该公司明确规定，广告不得包含性图像，包括裸露、性行为或性暗示内容。
Meta’s broader policies on adult sexual exploitation say that nonconsensual intimate imagery is not permitted and the promotion of “nudify” or “AI kissing” apps are not allowed. However, ads for nudify or undressing apps have been appearing on Meta’s platforms frequently over the past few years, with researchers finding thousands the company had failed to detect—by the company’s own count, it also removed 344,000 ads on Facebook and Instagram that were promoting the services between November last year and January and has filed a lawsuit against at least one firm promoting nudify services.
Meta 关于成人性剥削的更广泛政策规定，不允许存在非自愿私密影像，也不允许推广“脱衣化”或“AI 接吻”类应用。然而，过去几年里，针对此类应用的广告频繁出现在 Meta 平台上，研究人员发现了数千个该公司未能检测到的案例。据该公司自行统计，去年 11 月至今年 1 月期间，它在 Facebook 和 Instagram 上删除了 34.4 万条推广此类服务的广告，并已对至少一家推广“脱衣化”服务的公司提起诉讼。
“Sexual exploitation, including promoting nudify apps, is horrific, and we work aggressively to fight it,” says Cindy Southworth, head of women’s safety at Meta. “This is a highly adversarial space where bad actors constantly evolve their tactics, which is why we’re continually strengthening our defenses: removing over 340,000 violating ads for nudify apps late last year alone, launching new AI technology to better detect and block new violating ads, and taking legal action against nudify app developers.” Southworth says that Meta shares the information of apps and developers that it removes from its platform with other tech companies as well.
“包括推广‘脱衣化’应用在内的性剥削行为是可怕的，我们正在积极打击它，”Meta 女性安全主管辛迪·索斯沃斯（Cindy Southworth）表示。“这是一个高度对抗的领域，恶意行为者不断演变其策略，这就是为什么我们不断加强防御：仅去年年底就删除了超过 34 万条违规的‘脱衣化’应用广告，推出了新的 AI 技术以更好地检测和拦截新的违规广告，并对‘脱衣化’应用开发者采取法律行动。”索斯沃斯表示，Meta 还会与其他科技公司共享其从平台上移除的应用和开发者的信息。
Most of the video ads had a cover photo of something like a beach, mountains, or a waterfall, disguising the fact that the rest of the content was pornographic. “This is an example of an adversarial tactic these advertisers use to try and evade detection. We’re aware of this tactic and block and remove many violating ads despite them trying to obscure violating content with a benign image,” says Southworth. “That said, no system is perfect, and we’re continuing to investigate why these ads were [able to run].”
大多数视频广告的封面图是海滩、山脉或瀑布等，掩盖了其余内容为色情信息的事实。“这是这些广告主试图逃避检测所使用的对抗性策略的一个例子。我们了解这种策略，尽管他们试图用无害的图像掩盖违规内容，我们还是拦截并删除了许多违规广告，”索斯沃斯说。“话虽如此，没有完美的系统，我们正在继续调查这些广告为何（能够投放）。”