内容来源：https://www.wired.com/story/san-francisco-demands-apple-and-google-delete-ai-nudify-apps-from-app-stores/

内容总结：

美国旧金山要求苹果和谷歌下架“脱衣”AI应用，指其助长色情深度伪造并牟利

据《连线》杂志获取的勒令停止函显示，美国旧金山市检察官邱信福已于周四向苹果（Apple）和谷歌（Google）发出法律通知，要求这两大科技巨头从其应用商店中移除13款“换脸”应用程序。这些应用允许用户利用AI技术，在未经他人同意的情况下生成其裸体图像。

邱信福在接受采访时表示，生成非自愿的私密图像是“非法、有害且完全不可接受的”。他指出，苹果和谷歌很可能已从这些提供“脱衣”功能的应用中赚取了“数百万美元的费用”，并要求它们停止“帮助和教唆”销售露骨的深度伪造图像，并与相关应用开发者“断绝”业务关系。

“这些公司有责任确保其平台上的应用不会助长性虐待，”邱信福在函件中强调。函件指出，加州法律禁止支持创建深度伪造色情内容的服务，而这些应用通过应用内购方式获利，科技公司从中抽取分成。“全球最大、最老牌的科技公司竟然在纵容这种行为，这必须停止。”

研究人员的报告多次发现，苹果App Store和谷歌Play Store中存在可让用户利用AI生成色情图像的应用，其中一些应用的适龄评级甚至允许儿童使用。尽管新的法律和禁令试图解决这一网络毒瘤，但科技和社交媒体公司仍在持续引导数以百万计的用户接触这一有害技术。

对此，谷歌发言人表示，公司已因违反政策删除了“数百款”具备“脱衣”功能的应用，包括邱信福办公室点名的五款安卓应用，并限制了相关搜索词。苹果发言人则表示，开发者需对其应用内容负责，公司已下架了被点名的三款应用，并正在终止其开发者账户，其余四家开发者若未解决违规问题也将面临下架风险。

近五年来，高度盈利的深度伪造“脱衣”技术在网上泛滥。大量应用、网站和机器人允许用户（主要是男性）上传他人（绝大多数为女性及女孩）照片，通过数字手段“移除”其衣物或将其置入露骨性场景。只需一张参考照片和几次点击，有时只需几秒钟即可生成色情深度伪造内容。随着生成式AI技术的进步，这些图像和视频变得愈发逼真，甚至有人在至少90所学校中利用该技术制作了未成年人的深度伪造性虐待图像。

邱信福警告称：“这些图像被用来欺凌、羞辱和威胁女性及女孩。该行业对受害者的声誉、心理健康和自主权造成了可怕的影响，甚至有受害者因此产生自杀倾向。”

据旧金山市检察官办公室调查，这13款应用（其中8款在苹果商店，5款在谷歌商店）多以“换脸工具”自居，用户使用后即可解锁其创建色情深度伪造内容的功能。其中一款下载量超过100万的应用网页，展示了十多种AI生成图像风格，包括“比基尼女王曲线”、“恬静丰腴”和“电影感亲密”等，其中不乏女性色情化图像。

独立监督机构“技术透明度项目”在今年1月和4月的报告中，已发现两大应用商店中存在约100款此类应用。研究人员预估，这些被查明的应用总下载量约达4.8亿次，可能创造了约1.2亿美元的总收入。该机构主管凯蒂·保罗指出：“苹果和谷歌在营销中承诺其应用商店值得信赖且安全，但现实情况却并非如此。”

康奈尔大学和乔治城大学5月发布的一份研究预印本也显示，他们测试了155款通用“换脸”应用，其中70%的应用可以用于创建裸体图像换脸，且应用并未包含安全措施来预防这一行为。研究人员指出，许多AI图像生成及编辑应用实际上是“双用途”工具：它们表面上显得无害以规避平台内容审核，但实则具备创建有害内容的能力。

邱信福表示，他的办公室将持续追查这一问题，并力促苹果和谷歌立即下架这些应用并加强筛查系统。“我们希望这些公司能做正确的事——但如果它们不这样做，我们将不得不考虑所有法律选项。”

中文翻译：

据《连线》杂志查阅的发送给苹果和谷歌的停止侵权函显示，这两家公司已被要求下架能够“一键脱衣”或“裸体化”他人的应用程序，并必须停止从这一有害技术中获利。
本周四，旧金山市检察官戴维·邱向苹果和谷歌发出法律通知，要求其从应用商店中移除13款“换脸”应用。这些应用允许用户利用人工智能生成未经同意的裸体图像。函件称，硅谷巨头应停止“协助和教唆”销售露骨的深度伪造图像，并与相关应用开发商“断绝”商业关系。
“生成未经同意的私密图像是非法、有害且完全不可接受的行为。”邱告诉《连线》杂志。其办公室此前已对16个热门深度伪造网站采取法律行动。他表示，苹果和谷歌很可能从提供“脱衣”功能的应用中“赚取了数百万美元费用”，并应改进审核流程，从源头上阻止此类应用上架。
“这些公司有责任确保其平台上的应用不会助长性虐待行为。”邱表示。旧金山市的法律函件指出，加州法律禁止支持制作深度伪造色情内容的服务。函件称，这些应用通过内购收费，科技公司从中抽成。“全球最大、最成熟的科技公司竟然助长这种行为，必须予以制止。”
研究人员已多次发现并举报苹果App Store和谷歌Play Store中存在可通过AI生成性图像的应用——部分应用甚至被标注为适合儿童使用。尽管新法律和禁令试图解决网络上露骨深度伪造内容的祸害，科技和社交平台仍在持续将数百万人导向有害技术。
苹果和谷歌均有开发者政策禁止在其平台上出现色情、虐待和骚扰内容。此前在研究人员和记者曝光后，两家公司已移除数十款“脱衣”和深度伪造应用。
谷歌发言人丹·杰克逊告诉《连线》杂志，该公司已因违反政策删除了“数百款”具有脱衣功能的应用，包括邱办公室标记的5款安卓应用，并采取其他措施限制用户访问。
“谷歌Play不允许含有色情内容的应用，我们持续主动检测并移除有害内容的应用。”杰克逊在声明中表示，“当违规行为被举报时，我们会调查并迅速采取行动，包括暂停数百款违规应用、限制商店中‘脱衣’等搜索词。”
苹果则向《连线》杂志表示，开发者最终需为其应用内容负责，违反规则的开发者将被下架应用。该公司还表示用户可自行举报潜在问题应用。苹果发言人亚当·德马在声明中告诉《连线》杂志，已下架旧金山市检察官标记的3款应用，并“正从其开发者计划中终止相关账户”。另外4家开发者必须“解决政策违规问题，否则也面临下架风险”。
“App Store旨在为用户提供安全可信赖的环境，我们始终严格禁止设计用于生成、传播或消费色情内容的应用。”德马在声明中说，“‘脱衣’应用违反我们的应用审核指南，我们已主动拒绝许多此类应用并下架其他应用，包括用户通过举报工具标记的内容。”
过去五年，暴利的深度伪造“脱衣”技术在网上泛滥——最典型的是xAI公司Grok模型在1月份被用于生成数百万张色情化图像。大量应用、网站和机器人允许用户（多为男性）上传他人（绝大多数为女性及女童）照片，通过数字化手段“移除”衣物或将其置于露骨性场景中。
通常只需一张参考照片和几次点击即可生成色情深度伪造内容，部分结果数秒内即可呈现。随着底层生成式AI技术的进步，图像和视频愈发逼真，部分免费或低价即可生成有害内容。《连线》杂志与Indicator Media此前报道，至少90所学校发生过利用深度伪造生成未成年人性虐待图像的事件。
“这些图像被用于欺凌、羞辱和威胁女性及女童。”邱表示，“该行业对个人声誉、心理健康、自主权造成可怕影响。已有受害者产生自杀倾向。”
旧金山市检察官办公室调查的13款应用——其中8款在App Store、5款在Play Store——普遍以“换脸”工具为宣传，用户使用后才能发现其具备生成色情深度伪造内容的能力。一款下载量超百万的应用官网显示十几种AI图像风格，包括“比基尼女王曲线版”“丰满平静版”“电影级亲密版”。多款风格配有色情化女性图片及描述。另一款被标记应用的首页声称可生成“免费无审查”视频。《连线》杂志未公开应用名称以避免引导用户下载。
解决该问题对苹果和谷歌而言并不意外。过去一年，多份报告指出两家平台存在可生成未经同意的裸体图像或视频的应用。今年1月和4月，独立监督组织“技术透明度项目”（TTP）在App Store和Play Store发现约100款相关应用及部分脱衣技术广告。（谷歌的杰克逊称已下架TTP识别的大多数应用。）据估算，被识别应用总下载量约4.8亿次，总收入约1.2亿美元。
“我们以为首次报告后问题将不再出现——但第二次报告显示情况同样严重，甚至更糟。”TTP总监凯蒂·保罗表示，“苹果和谷歌在营销中大力宣传其应用商店的可信度和安全性，但现实并非如此。”
与此同时，康奈尔大学和乔治城大学研究人员在5月发布的预印本论文中，在谷歌和苹果应用商店识别出420款提供普通换脸功能的应用。他们测试了155款能否用于裸体图像换脸；其中70%可行，且应用未设置安全措施阻止此类使用。
“这些应用均未宣传为脱衣应用。”论文称，“这表明换脸应用及许多其他AI图像生成编辑应用实际上是‘双重用途’：它们因表面无害而规避平台内容审核，但具备生成有害内容的能力。”
旧金山市检察官邱表示，其办公室在“彻底震惊”于该技术的危害和规模后，将持续追究问题。“我希望苹果和谷歌立即下架这些应用、强化筛查系统，确保此类应用未来永不上架。”他说，“我们期待这些企业做出正确选择——但若不然，我们将考虑一切法律手段。”
（更新于美国东部时间2026年7月17日下午4:10：补充苹果发言人评论。）

英文来源：

Apple and Google have been ordered to take down apps that can “nudify” or “undress” people and told that they must stop profiting from the harmful technology, according to cease-and-desist letters sent to the companies seen by WIRED.
On Thursday, San Francisco city attorney David Chiu sent legal notices to Apple and Google demanding that they remove from their app stores 13 face-swapping apps, which allow users to create AI-generated nonconsensual nude images. The letters say the Silicon Valley giants should stop “aiding and abetting” the sale of explicit deepfake images and “sever” business relationships with the app developers.
“Generating non-consensual intimate images is illegal, harmful, and completely unacceptable,” Chiu tells WIRED. The city attorney, whose office previously took legal action against 16 popular deepfake websites, says Apple and Google have likely “made millions of dollars in fees” from apps that offer nudification, and they should improve their moderation processes to stop them appearing in their stores in the first place.
“These companies have responsibility to ensure that apps on their platforms do not facilitate sexual abuse,” Chiu says. The city’s legal letters say California’s laws prohibit supporting services that create deepfake pornography. The apps use in-app payments, which the tech companies take a cut of, the letters says. “The fact that some of the world’s largest and most established technology companies are facilitating this has to stop.”
Researchers have repeatedly found and reported apps in Apple’s App Store and Google’s Play Store that allow people to generate sexual images using AI—including some apps being rated as suitable for use by children. While new laws and bans aim to tackle the scourge of explicit deepfakes online, technology and social media companies consistently direct millions of people toward the harmful tech.
Both Apple and Google have developer policies that prohibit pornography, abuse, and harassment on their platforms. They have previously removed dozens of nudify and deepfake apps, after reports by researchers and journalists.
Google spokesperson Dan Jackson tells WIRED that the company has deleted “hundreds” of apps with nudifying features for policy violations, including the five Android apps flagged by Chiu’s office, among other steps to restrict access to them.
“Google Play does not allow apps that contain sexual content, and we continually take proactive steps to detect and remove apps with harmful content,” Jackson says in a statement. “When violations are reported to us, we investigate and take swift action, which in the case of these apps has included suspending hundreds of violating apps and restricting related search terms like 'nudify' on our store."
Apple tells WIRED that developers are ultimately responsible for the content of their apps, and those who fail to follow the rules will have their apps removed. The company also says users can report potentially problematic apps themselves. Adam Dema, an Apple spokesperson, tells WIRED in a statement that the company has removed three of the apps flagged by San Francisco city attorney and is “in the process of terminating their developer accounts from our program.” The other four developers must “address policy violations or risk being removed as well,” the company says.
“The App Store was designed to be a safe and trusted place for users, and we have always strictly prohibited apps designed to generate, distribute, or consume pornography," Dema says in the statement. “‘Nudification' apps are against our App Review Guidelines and we have proactively rejected many of these apps and removed many others, including when users have flagged them via our reporting tools.”
Over the last five years, a highly lucrative slurry of deepfake “nudification” tech has emerged online—most transparently with xAI’s Grok being used to create millions of sexualized images in January. A host of apps, websites, and bots allow people (mostly men) to upload pictures of people (overwhelmingly women and girls) and digitally “remove” clothing or place them into graphic sexual scenarios.
Often all it takes to create sexual deepfakes is a reference photo and a couple of clicks, with some results available in seconds. Images and videos have become more realistic as the underlying generative AI technology has improved, with services providing some results for free or charging small fees to create the harmful content. Previous reporting by WIRED and Indicator Media has uncovered incidents in at least 90 schools where deepfake sexual abuse images have been created of minors.
“These images are used to bully, humiliate, and threaten women and girls,” Chiu says. “This industry has a horrific impact on one’s reputation, mental health, loss of autonomy. There have been victims who’ve been suicidal.”
The 13 apps investigated by the City Attorney’s Office—eight on the App Store and five on the Play Store—broadly advertise themselves as “face-swapping” tools, with their ability to create sexual deepfakes available once people use them. The website of one app, which has more than 1 million downloads, displays more than a dozen different styles of AI images it will generate, including “bikini queen curvy,” “calm busty,” and “cinematic intimacy.” Many of the styles show sexualized images of women alongside their descriptions. The homepage of another of the targeted apps claims to produce “free and uncensored” videos. WIRED is not naming the apps to avoid pushing people toward them.
The problem won’t come as a surprise to Apple and Google. Over the last year, multiple reports have identified apps on the companies’ platforms that can allow people to create nonconsensual nude images or videos. In January and April this year, the Tech Transparency Project, an independent watchdog group, uncovered around 100 apps across both the App Store and Play Store, as well as some advertisements for nudifying technology on the platforms. (Google’s Jackson says the company has removed most of the apps identified by TTP.) Apps identified by the research were estimated to have been collectively downloaded around 480 million times and may have made around $120 million in combined revenues.
“We didn’t think after the first report that we would see this as a problem again—and it was just as bad, if not worse, after the second report,” says Katie Paul, the director of TTP. “Apple and Google make a lot of promises in their marketing about how trusted and safe their app stores are. And that is just not what is playing out in reality.”
Meanwhile, in a preprint research paper published in May, researchers from Cornell University and Georgetown University identified 420 apps offering general face-swapping capabilities on Google’s and Apple’s app stores. They tested 155 to see if they could be used to create face swaps with nude images; in 70 percent of cases, it was possible, with the apps not including safety measures to prevent this.
“None of these apps are advertised as nudification apps,” the research says. “This suggests that face swap apps, and many other forms of AI image generation and editing apps, are effectively ‘dual-use’: apps that evade content moderation by platforms because they present as benign, but possess the capability to create harmful content.”
Chiu, the San Francisco city attorney, says his office will keep pursuing the problem after being “absolutely horrified” at the harm and scale of the technology. “My hope is that Apple and Google will immediately remove these apps and strengthen their screening systems to make sure that apps like this never get onto their platforms in the future,” he says. “It's our hope that these companies will do the right thing—but if they don’t, we will have to consider all of our legal options.”
Updated 4:10 pm ET, July 17, 2026: Added comment from an Apple spokesperson.