YouTube和X已成为“去衣”应用的“入口”

qimuai 发布于 阅读:47 一手编译

YouTube和X已成为“去衣”应用的“入口”

内容来源:https://www.wired.com/story/youtube-and-x-have-become-gateways-to-nudify-apps/

内容总结:

主流社交平台成“一键脱衣”软件引流渠道,报告揭示惊人数据

据反极端主义组织“战略对话研究所”(ISD)最新发布的报告显示,尽管主流社交媒体平台均设有严格规定,禁止传播未经同意的私密影像(NCII),但许多平台实际上却直接向用户推荐此类内容。该报告于周一发布,重点分析了排名前十的“一键脱衣”应用程序和网站,并追踪了用户发现这些工具的途径。

研究发现,令人惊讶的是,大量流量并非来自监管松散的小众网络社区(如4chan),而是来自主流社交媒体平台。在2025年12月至2026年3月期间,社交网络为这些“脱衣”网站带来了超过570万次访问量。其中,YouTube是最大的流量来源,贡献了182万次访问,占比超过30%。这些视频通常通过搜索“脱衣应用”等关键词找到,内容涵盖应用评测、推广甚至提供免费积分的促销代码。X平台(原推特)紧随其后,带来了超过130万次访问。

报告作者指出,这一发现与YouTube禁止色情内容的政策“直接冲突”。理论上,其政策应涵盖生成非自愿色情图像的“脱衣”网站或工具,但违反规定的视频却极易在平台上被发现和访问,使平台成为了通往这些网站的“门户”。ISD高级研究主任梅兰妮·史密斯表示,YouTube不仅是流量的“被动来源”,在很多情况下,它还在“积极促成”这些工具的使用。尽管YouTube发言人声称其有严格政策禁止包括非自愿私密影像在内的不良内容,但史密斯认为,平台对禁止外部链接或广告的规定执行并不彻底。

报告还揭示了“脱衣”软件的定价与普及程度。用户最低仅需1美元即可生成一张色情图片,而这类平台年收入总额可能高达3600万美元。这些应用的目标对象通常包括现任或前任女友,甚至令人不安地包括姐妹、表亲等亲属。更令人担忧的是,史密斯指出,许多用户的使用动机并非“出于性目的”,而是“为了让受害者丢掉工作,或以恶意方式损害其生计与生活”。

此类AI生成图像的泛滥已成为社会顽疾。例如,2026年1月,X平台因其AI聊天机器人Grok在未经同意的情况下生成女性裸照(包括未成年人)而遭到强烈批评。尽管美国联邦《Take It Down Act》已全面生效,要求平台在收到受害者删除请求后48小时内移除相关影像,明尼苏达州也已率先立法禁止“脱衣”应用,但此类软件的传播势头不减,且更加易于获取和使用。ISD报告呼吁采取全系统协同应对措施,包括加强平台监管和资助校园数字素养教育。然而,鉴于近期调查发现全球已有超过90所学校报告深度伪造相关案例,这一问题的蔓延短期内恐难遏制。

中文翻译:

多数主流社交媒体平台都制定了严格的规定,禁止传播未经同意的私密影像(NCII)或露骨的色情照片和视频。但一份新报告发现,许多社交媒体平台会直接引导用户访问此类内容。
反极端主义组织“战略对话研究所”(ISD)发布的这份报告,分析了助长所谓“脱衣”应用和网站泛滥的网络生态。这些应用和网站能让用户未经他人同意,以数字方式“剥去”其衣物。
这项周一公布的研究调查了制作未经同意的露骨深度伪造内容的前十大应用和网站,并探究了人们是如何找到它们的。研究发现,令人惊讶的是,大量引流并非来自4chan等监管不严的小型网络社区,而是来自主流社交媒体平台。
根据ISD的报告,在2025年12月至2026年3月期间,社交网络为“脱衣”网站带来了超过570万次访问。其中最大的流量来源是YouTube,贡献了182万次访问,占引流量的30%以上。
这些视频源于对“脱衣应用”或“脱衣软件”等关键词的搜索,内容包括测评和推广特定应用,以及提供链接和优惠码来获取免费积分。研究显示,X平台是这些网站的第二个主要流量来源,贡献了超过130万次访问。
研究报告的作者写道,这些发现似乎与YouTube禁止露骨色情内容的规定“直接冲突”。“按理说,这应该包括生成未经同意的露骨图像的‘脱衣’网站或工具,”报告称,“然而,违反这些规定的内容在该平台上很容易被发现和访问,这实际上使其成为了通往‘脱衣’网站的门户。”
ISD研究与政策高级总监梅兰妮·史密斯告诉《连线》杂志:“YouTube不仅仅是引流的被动来源。在许多情况下,它也在协助这些工具的使用。”
有趣的是,史密斯指出,YouTube的规定不仅禁止发布露骨色情内容,还禁止发布指向或宣传露骨色情网站的链接(例如,链接到OnlyFans)。“理论上,这应该涵盖未经同意的图像、色情报复或裸照泄露,但似乎并未得到全面执行,”她说。
在回应《连线》杂志的置评请求时,YouTube发言人布特·布尔温克尔表示,公司有“严格的政策,禁止包含非自愿性化内容,如未经同意分享的私密影像。”他指出,这些政策既适用于YouTube自身的内容,也适用于外部链接,并包括“逼真模拟裸体的篡改或合成内容。”
该研究还考察了“脱衣”工具的定价和可获取性,一些应用和网站允许用户仅花1美元就能生成一张露骨的色情图片。尽管使用这些平台的成本相对较低,但它们可能利润极高,最近《连线》杂志的一篇报道发现,它们每年可能总共产生高达3600万美元的收入。
研究发现,“脱衣”应用的常见目标包括现任和前任女友,以及(令人不安地)姐妹和表亲等亲属。作者还惊讶地发现,使用“脱衣”工具的人的动机甚至不一定“与性有关”,史密斯说:“许多请求是关于让人丢掉工作,以恶意方式损害他们的生计和生活。”
“脱衣”应用已成为许多社交媒体平台上的一个主要祸害。然而,有时平台不仅允许这些AI生成的图像传播,还会帮助人们制作它们。
2026年1月,X平台因人们开始使用AI聊天机器人Grok,未经同意生成女性(包括一些未成年人)的裸体或性暗示图像而受到强烈批评。这场风波最终促使该公司发表声明,称将限制Grok仅向付费用户开放。“我们仍然致力于让X成为每个人的安全平台,并继续对任何形式的儿童性剥削、未经同意的裸体以及不受欢迎的色情内容采取零容忍态度,”声明写道。X平台未回应置评请求。
在美国,NCII是非法的。联邦《下架法案》于今年5月全面生效,规定社交媒体平台必须在受害者提交下架请求后的48小时内删除未经同意传播的图像。大多数州已通过了某种形式的反深度伪造立法,2026年5月,明尼苏达州成为首个专门禁止“脱衣”应用的州。
然而,尽管有这类立法,这些应用仍在不断扩散,而且变得更加易于获取和使用。
ISD的报告强调需要“采取协调一致、覆盖整个系统的应对措施,涉及线上、线下、混合领域及政策干预”,例如加强平台监管和为学校数字素养讲习班提供资金。但鉴于《连线》杂志最近的一项调查发现,全球已有超过90所学校报告了深度伪造案例,“脱衣”应用的泛滥似乎短期内不会减缓。

英文来源:

Most mainstream social media platforms have strict guidelines preventing the dissemination of nonconsensual intimate images (NCII), or sexually explicit photos and videos. But a new report finds that many social media platforms directly refer users to such content.
The report from the Institute for Strategic Dialogue (ISD), an anti-extremism organization, analyzes the online ecosystem that enables the proliferation of so-called “nudify” apps and websites, which allow users to digitally undress people without their consent.
The study, which was published on Monday, looked at the top 10 apps and websites used to make nonconsensual explicit deepfakes, as well as how people are finding them. It found that a surprising amount of referral traffic comes not from small, poorly regulated online communities like 4chan but from mainstream social media platforms.
According to the ISD report, social networks drove more than 5.7 million visits to nudify sites between December 2025 and March 2026. The top driver of this traffic was YouTube, which was responsible for 1.82 million site visits, more than 30 percent of the referrals.
Such videos, which resulted from searches for keywords like “undress app” or “nudify app,” ranged from reviewing and promoting specific apps to linking to promo codes to provide free credits. X was the second most prominent source of traffic to the sites, accounting for over 1.3 million visits, according to the study.
The study’s authors wrote that these findings appear to be in “direct conflict” with YouTube’s policies, which prohibit sexually explicit content. “This should logically include nudify websites or tools that generate nonconsensual explicit imagery,” the report reads. “However, content violating these policies was easily discoverable and accessible on the platform, effectively turning it into a gateway to nudify websites.”
“It wasn’t just that YouTube was a passive source” of referral traffic, Melanie Smith, the senior director of research and policy for ISD, tells WIRED. “In a lot of these cases, it was facilitating the use of these tools as well.”
Interestingly, Smith notes, YouTube’s policies don’t just prohibit posting sexually explicit content, they also prohibit posting links to or advertising for sexually explicit websites (say, linking to an OnlyFans). “In theory that should cover nonconsensual imagery and revenge porn, or nude photo leaks, but it doesn’t seem like that’s being enforced comprehensively,” she says.
In response to WIRED’s request for comment, YouTube spokesperson Boot Bullwinkle says the company has “strict policies prohibiting content that includes unwanted sexualization, such as nonconsensually shared intimate imagery.” He noted that these policies apply both to content on YouTube itself and to external links, and include “altered or synthetic content that realistically simulates nudity.”
The study also looked at the pricing and accessibility of nudification tools, with some apps and websites allowing users to generate sexually explicit content for as little as $1 an image. Despite the relatively low cost of using the platforms, they can be extremely profitable, with a recent WIRED report finding they may be generating as much as $36 million in collective revenue per year.
Common targets of nudification apps include current and ex-girlfriends, as well as (disturbingly) relatives such as sisters and cousins, the study found. The authors were also surprised to find that the motivations of people who use nudify tools are not even necessarily “sexual in nature,” says Smith: “A lot of the requests were about getting people fired from jobs and compromising their livelihoods and lives in nefarious ways.”
Nudify apps have become a major scourge on many social media platforms. Sometimes, however, a platform will not only allow these AI-generated images to spread but also help people make them.
In January 2026, X was subject to intense criticism when people started using the AI chatbot Grok to generate nude or sexually suggestive images of women without their consent, including some minors. The uproar eventually prompted the company to issue a statement saying it was limiting access to Grok to paid users. “We remain committed to making X a safe platform for everyone and continue to have zero tolerance for any forms of child sexual exploitation, non-consensual nudity, and unwanted sexual content,” the statement read. X did not respond to a request for comment.
NCII is illegal in the United States, with the federal Take It Down Act, which went into full effect in May, mandating that social media platforms must remove nonconsensually distributed images within 48 hours of a victim submitting a takedown request. Most states have adopted some form of anti-deepfake legislation, and in May 2026, Minnesota became the first state to ban nudification apps specifically.
Despite such legislation, however, the apps have only continued to spread—and they’ve also gotten more accessible and easier to use.
The ISD report emphasizes the need for “coordinated, system-wide responses spanning online, offline, hybrid, and policy interventions,” such as increased platform regulation and funding for school digital literacy workshops. But with a recent WIRED investigation finding that there have been reported deepfake cases in more than 90 schools around the world, it doesn’t seem like the proliferation of nudify apps is slowing down anytime soon.

连线杂志AI最前沿

文章目录


    扫描二维码,在手机上阅读