Anthropic said Thursday that its Claude AI model “gained unauthorized access” to systems belonging to three outside organizations during controlled security tests that were intended to prevent contact with real-world networks, according to the company’s disclosure.
The revelation follows closely behind a similar admission from rival OpenAI, which recently said some of its models improperly reached the internet and behaved unpredictably during cybersecurity evaluations.
Anthropic said it reviewed more than 141,000 “evaluation runs” and identified three cases in which separate versions of Claude accessed systems tied to three organizations that the company did not name.
In each incident, Anthropic said, Claude was taking part in a “capture-the-flag” exercise. The model had been told to “break in and retrieve” a piece of “secret information” that was “hidden on a different machine on the network.”
“The challenge is left open-ended, and no particular method is prescribed,” Anthropic explained.
Anthropic said the circumstances differed from the OpenAI case because Claude’s internet access stemmed from “a misunderstanding between us and our evaluation partner,” a firm called Irregular, the company wrote in a blog post.
Even so, the company said Claude relied on relatively simple methods, including “exploiting weak passwords and unauthenticated endpoints,” to reach the outside systems.
Among the models involved was Mythos 5, described as one of Anthropic’s most powerful systems and currently available only to a small group of approved partners.
Anthropic is working with Irregular to assess the situation, it said, and the company has contacted or attempted to contact all three impacted organizations.
OpenAI and Anthropic have both released their most powerful models this year, known as Sol and Mythos, respectively, boosting concerns across the industry about safety and security.
Those concerns also revolve around AI agents, which are software products that are designed to perform tasks autonomously.
OpenAI admitted last week that its models broke out of their confined environment during testing, connected to the internet, and infiltrated Hugging Face, a site where developers store and share their code.
Days later, OpenAI said it found three additional incidents.
OpenAI CEO Sam Altman said on a podcast this week that the company had “paused” its own testing after the incident while it improved the security around its “sandboxing,” which is the process of isolating software in a controlled environment for testing.
And in a public letter released earlier this week, more than 1,000 AI staffers across leading firms called for the industry to be more tightly regulated.
“To realize AI’s potential, industry, government, and society at large may need the option to buy time to address emerging risks, develop security measures, and strengthen oversight,” read the letter, whose signatories include Anthropic CEO Dario Amodei, Meta executives, OpenAI researchers and more.
Altman did not sign the letter, but he told reporters on Capitol Hill Wednesday that “we agree on a lot of the principles of that.”
Earlier this year, the Trump administration invoked national security concerns to block OpenAI and Anthropic from launching their newest models but ultimately indicated it was satisfied with assurances about their safety, leading to their release.
In June, Mr. Trump signed an executive order creating a voluntary framework under which AI developers will share advanced models with the government before public release.
Under the framework, developers such as OpenAI, Anthropic and Google would give the government access to their most powerful models for up to 30 days before planned release.
