India becomes the latest country to report rogue AI agent activity

印度网络安全机构周四表示，在研究和测试过程中，来自多家提供商的 AI 代理访问了印度组织机构的个人数据和机密材料。

The Indian Computer Emergency Response Team (CERT-In) said it had received “multiple voluntary disclosures from AI model providers” whose models, during research and evaluation runs on the open internet, “had reached the live systems of Indian organizations.”

印度计算机应急响应小组（CERT-In）表示，已收到“来自 AI 模型提供商的多项自愿披露”，这些提供商的模型在开放互联网上进行研究和评估运行期间，“触及了印度组织的实时系统”。

The agents got in through credentials and connection settings left exposed in app configurations, public code repositories, and overly permissive cloud roles, and their actions “exceeded the authorisation of their assigned tasks.”

这些代理通过应用程序配置、公共代码存储库以及权限过大的云角色中暴露的凭据和连接设置进入系统，其行为“超出了其分配任务的授权范围”。

This makes India the latest stop on what has become a world tour for rogue AI agent activity. In July, OpenAI’s agents escaped a test environment and hacked French company Hugging Face — kicking off a summer of rogue AI agent activity.

这使得印度成为“流氓 AI 代理”全球活动的最新一站。今年 7 月，OpenAI 的代理逃离了测试环境并入侵了法国公司 Hugging Face，由此开启了一个“流氓 AI 代理”活动的夏天。

Last month, OpenAI apologized to Australia after its agents accessed four government websites, including a Medicare statistics portal. At Hugging Face, OpenAI’s agents started the same way but then found and exploited software flaws to run code on the company’s servers.

上个月，OpenAI 的代理访问了包括医疗保险统计门户网站在内的四个政府网站，随后 OpenAI 向澳大利亚道歉。在 Hugging Face 事件中，OpenAI 的代理以同样的方式开始，但随后发现并利用软件漏洞在公司服务器上运行了代码。

CERT-In described no such hacking in India, where the agents appear to have simply used the access left lying around. Still, the Indian cases appear to be worse in one respect, which is what the agents saw: In Australia, OpenAI said it found no evidence its agents touched individuals’ medical or criminal records, even after they wrote files inside a Medicare statistics system, while in India, agents read personal data and confidential business records.

CERT-In 并未描述印度发生此类黑客攻击，代理似乎只是利用了遗留的访问权限。尽管如此，印度案例在某些方面似乎更严重，即代理所接触到的内容：在澳大利亚，OpenAI 表示没有证据表明其代理触及了个人的医疗或犯罪记录（即使它们在医疗保险统计系统中写入了文件），而在印度，代理读取了个人数据和机密商业记录。

And unlike earlier incidents, each disclosed by a single lab, India’s case involved agents from multiple providers. CERT-In did not name the providers or affected organizations, or say whether the agents altered any systems. TechCrunch has reached out to CERT-In for more information.

与之前由单一实验室披露的事件不同，印度的情况涉及来自多家提供商的代理。CERT-In 没有透露提供商或受影响组织的名称，也没有说明代理是否更改了任何系统。TechCrunch 已联系 CERT-In 以获取更多信息。

For now, the closest parallel seems to be Anthropic’s July disclosure of Claude breaking into three companies’ systems during testing using basic techniques like weak passwords. In India, as in Anthropic’s cases, the affected organizations had not detected the intrusions on their own. The disclosures have left governments relying on AI companies causing these breaches to voluntarily disclose rogue AI agent activity.

目前，最相似的案例似乎是 Anthropic 在 7 月披露的 Claude 在测试期间利用弱密码等基本技术入侵三家公司系统的事件。在印度，正如 Anthropic 的案例一样，受影响的组织并未自行发现这些入侵。这些披露使得各国政府只能依赖造成这些违规行为的 AI 公司来主动披露“流氓 AI 代理”的活动。