Anthropic 称 Claude 在网络安全测试中入侵了三家机构

美国科技公司 Anthropic 表示，在一项私密安全实验期间，其 AI 模型自行入侵了三家机构的系统。

这些模型在一个本应隔离的测试环境中发现了漏洞，并连接到了互联网。

就在数日前，竞争对手 OpenAI 表示，其模型曾入侵其他公司的系统，其中包括 AI 工具平台 Hugging Face。

这一消息促使 Anthropic 检查其自身系统是否也实施过类似攻击。该公司称，已发现三起此类事件，并已向受影响公司报告。

Anthropic 未透露这些机构的名称，并敦促其他 AI 实验室进行类似审查，以更好地理解其模型能力带来的风险。

Anthropic 在一份声明中表示，公司审查了超过 14 万项测试，以寻找证据证明 Claude——其 AI 模型家族——是否曾在本应隔离、无法访问互联网的测试环境中成功上线。

这些测试包括要求 Claude 获取封闭网络中另一台机器上隐藏的“秘密”信息的演练。

随后，Claude 被要求通过入侵该机器并找到相关信息来获取这些内容——这是专家评估模型黑客攻击能力的常见方式。

Anthropic 及其测试合作伙伴运行的系统中存在一项“配置错误”，使模型获得了真实的互联网访问权限。

这家总部位于旧金山的公司称，Claude 将这一切仍视为同一场演练的一部分，随后连接互联网并入侵了三家真实机构的系统，而非仅攻击测试系统。

Anthropic 表示，最早的事件可追溯至 4 月，公司正“将修复工作视为完全由我们承担责任”来推进。

当时，Anthropic 和遭入侵的机构都未察觉这些入侵行为。

Anthropic 表示，本可以对记录进行更彻底的审查；同时，公司补充称，这些发现让其对通过更多投入和更严格的措施克服此类风险抱有“谨慎乐观”的态度。

US technology firm Anthropic says its AI models hacked into the systems of three organisations on their own, during a private security experiment.

The models found a weakness in what was supposed to be an isolated test environment and connected to the internet.

It comes just days after rival OpenAI said that its models had breached the systems of other companies, including AI tools hub Hugging Face.

The announcement prompted Anthropic to check whether its own systems had carried out similar attacks. It says it uncovered three cases which have since been reported to the affected companies.

Anthropic, which did not name the organisations, urged other AI labs to perform similar reviews to better understand the risks of their models' capabilities.

Anthropic said in a statement, external it reviewed more than 140,000 tests to find evidence Claude - its family of AI models - had managed to get online even though it was supposed to be in an isolated test environment, cut off from the internet.

The tests included exercises in which Claude was tasked with obtaining "secret" information hidden on another machine on the closed-off network.

It was then told to get the information by breaking into the machine and finding it - a common way that experts assess a model's hacking capabilities.

A "misconfiguration" on systems run by Anthropic and its testing partner left the models with live internet access.

Treating it all as still part of the same exercise, Claude then connected to the internet and breached the systems of three real organisations rather than just test ones, the San Francisco-based firm said.

Anthropic said the earliest incidents date back to April and that it is "approaching the fixes as if the responsibility were ours alone."

Neither Anthropic nor the organisations that were breached had noticed the intrusions at the time.

Anthropic said it could have reviewed its records more thoroughly and added that the findings gave the firm "cautious optimism" that such risks can be overcome with more investment and tighter measures.