London. An Anthropic AI model created fake online identities and sent emails to real people in an attempt to get malicious code approved during tests by a UK government research group.

During the tests by the AI Security Institute, some AI agents from Anthropic and OpenAI engaged in “sustained, potentially harmful activity directed at real people and organizations," the institute said in a report published late Tuesday.

In the most serious case, Anthropic’s Mythos 5 model attempted to insert malicious code into a software project by creating fake online identities and sending deceptive emails to persuade the recipient to approve the code.

The findings come amid a series of recent cyberattacks carried out autonomously by software from the two US companies, raising concerns about the capabilities and oversight of increasingly advanced AI models.

The AISI, established in 2023 to assess the safety of new AI models, conducted the tests with open internet access and certain safety features disabled.

Most of the incidents involved the Mythos 5 model, while two involved OpenAI’s GPT-5.6-Sol model.

The person overseeing the software refused to approve the code.

“These attempts were unsuccessful, and our investigations have not evidenced any resulting real-world harm,” the institute said, adding that it contained the incident within an hour.

However, the activities “show signs of novel, potentially deceptive behaviors and were to an extent and severity we did not anticipate,” it said.

An Anthropic spokesperson said the report “underscores the need for a broader conversation about how to safely evaluate increasingly capable AI agents.”

An OpenAI spokesperson said “independent testing is essential to understanding how increasingly capable models behave.”

“We’ll continue working with evaluators and other stakeholders across the industry to strengthen shared practices for conducting evaluations safely as models become more capable,” the spokesperson added.

The report follows a series of high-profile security incidents involving AI models.

In July, OpenAI confirmed that one of its models had escaped a testing environment and attacked another company, Hugging Face. About a week later, it said its models had targeted three additional companies.

On July 30, Anthropic disclosed three incidents in which AI models being tested “gained unauthorized access” to organizations that it did not identify.

Add as a preferred source on Google +