男子利用Grok生成7000张继女性虐图像后自杀身亡
英文标题：🔶 Man Dies by Suicide After Using Grok to Make 7000 Sexual Images of Stepdaughter
一项扩大的诉讼指控xAI的Grok聊天机器人被用于在数月内生成约7000张AI生成的儿童性虐待图像，涉及一名11岁女童，制作者被捕后自杀身亡。诉状称xAI的安全防护未能阻止长期滥用，且向国家失踪与被剥削儿童中心报告时信息不足，延误调查。此案凸显生成式AI被用于制作儿童性虐待材料的严峻趋势，对AI平台的安全责任提出重大挑战。
1.4 KB
1.1 KB
958 B
7.1 KB
4.8 KB
33.0 KB
7.3 KB
60.9 KB
3.2 KB
17.4 KB
14.9 KB
169.2 KB
142.7 KB
162.7 KB
11.1 KB
132.3 KB
206.7 KB
121.8 KB
229.3 KB
16.6 KB
中文
一项扩大的诉讼指控，一名男子利用xAI的Grok聊天机器人在数月内生成约7000张AI生成的儿童性虐待图像，受害者是一名11岁女童；该男子被捕后自杀身亡。
本周在联邦法院提交的修订诉状扩大了一起针对xAI的拟议集体诉讼的范围，指控该公司的安全防护未能阻止对Grok的长期滥用，且向当局的报告不充分。诉状称，继父的死亡使得本案被指受害者（化名Jane Doe 4）继续推进诉讼。
Grok被指控如何被滥用
根据修订诉状，化名Jane Doe 4的11岁女童的继父在数月间利用Grok，从该儿童的一张照片中生成了数千张露骨色情图像和视频。诉状称，这些材料包含乱伦和强奸的描绘。
诉状指出，此事曝光是因为该男子输入了一段提及“轮奸”的提示词，触发了xAI向国家失踪与被剥削儿童中心（NCMEC）提交的CyberTip举报。NCMEC将报告转交执法部门，后者随后取得搜查令并扣押了该男子的电子设备。
诉状称，法证检验据称发现了约7000张AI生成的图像和视频。诉状进一步指控xAI仅报告了原始照片，并未提供生成的图像或用户IP地址，从而延误了调查。
诉状援引的法庭记录显示，该男子被捕后获保释，两天后自杀身亡。
诉讼针对Grok的安全控制措施
修订诉状指控xAI在设计Grok时未配备足够的安全防护，未能防止利用真实儿童照片生成AI儿童性虐待材料。原告称，包括“Spicy Mode”在内的功能使该聊天机器人能够生成露骨图像，尽管安全防护本应阻止此类请求。
诉讼还指控xAI在提交CyberTip后通过隐瞒关键信息（包括AI生成的图像和用户IP地址）阻碍了调查。原告正在寻求损害赔偿以及一项要求加强安全防护的禁令。
xAI曾公开指出安全防护问题
xAI尚未对修订诉状公开回应。
此前，在一月份有报道称Grok生成了穿着极少衣物的未成年人图像后，该公司承认了Grok图像生成安全防护存在问题。
Grok的官方账号在X平台的一篇帖子中将此问题描述为“安全防护的疏漏”，并表示正在改进以阻止此类请求，同时补充称儿童性虐待材料非法且在平台上被禁止。
大约同一时间，xAI创始人埃隆·马斯克表示，他“不知道Grok生成过任何未成年裸体图像。完全是零。”他还表示，Grok不会生成非法内容，任何漏洞都会被迅速修复。
AI生成虐待报告激增
修订诉状提交之际，儿童保护机构和监管机构报告称，涉及生成式AI的案件急剧增加。
国家失踪与被剥削儿童中心表示，2025年已收到超过150万份与生成式AI相关的CyberTipline报告。
在剔除不含可操作信息的自动提交后，该组织称有超过7000份报告涉及用户生成或持有AI生成的儿童性虐待材料。
此案是针对xAI就Grok图像生成能力提起的多起法律行动之一。今年1月，来自美国35个州的检察长敦促该公司加强防护，防止生成未经同意的私密图像和儿童性虐待材料。
加利福尼亚州检察长罗布·邦塔也在有报道称Grok生成了儿童露骨图像后启动了调查。
English / 原文
An expanded lawsuit alleges xAI's Grok chatbot was used to create about 7,000 AI-generated child sexual abuse images of an 11-year-old girl over several months, and claims the man who made them later died by suicide after his arrest.
Filed this week in federal court, the amended complaint broadens a proposed class action against xAI, alleging the company's safeguards failed to stop prolonged misuse of Grok and that its reporting to authorities was inadequate. The filing says the stepfather's death left the alleged victim, identified as Jane Doe 4, to continue pursuing the case.
How Grok Was Allegedly Used
According to the amended complaint, the stepfather of an 11-year-old girl identified as Jane Doe 4 used Grok over several months to generate thousands of sexually explicit images and videos from a single photograph of the child. The filing alleges the material included depictions of incest and rape.
The complaint says the activity came to light after the man entered a prompt referring to a 'gang rape', triggering a CyberTip submitted by xAI to the National Center for Missing & Exploited Children (NCMEC). NCMEC referred the report to law enforcement, which later obtained a search warrant and seized the man's electronic devices.
A forensic examination allegedly uncovered about 7,000 AI-generated images and videos, according to the complaint. The filing further alleges xAI reported only the original photograph and did not provide the generated images or the user's IP address, delaying the investigation.
Court records cited in the complaint say the man was arrested, released on bail and died by suicide two days later.
Lawsuit Targets Grok's Safety Controls
The amended complaint alleges xAI designed Grok without adequate safeguards to prevent the creation of AI-generated child sexual abuse material from photographs of real children. Plaintiffs claim features including 'Spicy Mode' enabled the chatbot to generate explicit images despite safeguards that should have blocked such requests.
The lawsuit further alleges xAI hindered the investigation by withholding key information after filing its CyberTip, including the AI-generated images and the user's IP address. Plaintiffs are seeking damages and an injunction requiring stronger safeguards.
xAI Previously Flagged Safeguard Problems
xAI has not publicly responded to the amended complaint.
The company previously acknowledged problems with Grok's image-generation safeguards after reports in January that the chatbot produced images of minors in minimal clothing.
In a post on X, Grok's official account described the issue as 'lapses in safeguards' and said improvements were under way to block such requests, adding that child sexual abuse material is illegal and prohibited on the platform.
Around the same time, xAI founder Elon Musk said he was 'not aware of any naked underage images generated by Grok. Literally zero.' He also said Grok does not produce illegal content and that any bugs are fixed quickly.
AI-Generated Abuse Reports Surge
The amended complaint arrives as child protection agencies and regulators report a sharp rise in cases involving generative AI.
The National Center for Missing & Exploited Children said it received more than 1.5 million CyberTipline reports linked to generative AI in 2025.
After excluding automated submissions that contained no actionable information, the organisation said more than 7,000 reports involved users generating or possessing AI-generated child sexual abuse material.
The case is one of several legal actions filed against xAI over Grok's image-generation capabilities. In January, attorneys general from 35 US states urged the company to strengthen safeguards against the creation of non-consensual intimate images and child sexual abuse material.
California Attorney General Rob Bonta also opened an investigation following reports that Grok generated explicit images of children.