OpenAI admits that its models tried to access without permission websites of the government of the United States

BarcelonaArtificial intelligence agents developed by OpenAI attempted to gain unauthorized access to websites of United States Government agencies and a university, according to an investigation published this Friday by The New York Times. The artificial intelligence models did so without having received explicit instructions to do so, according to the company.

Among the identified targets are sites linked to the Securities and Exchange Commission (SEC) and the US Department of Commerce, from which the artificial intelligence accessed census data, as confirmed by OpenAI. The company is also investigating an access attempt on the Department of Education's website.

This incident is part of a series of unforeseen behaviors detected in OpenAI agents designed to perform search and information gathering tasks. In some cases, when the systems failed to obtain certain data through standard channels, they began to explore vulnerabilities in the pages to attempt to access restricted information, according to Efe.

The investigation adds to an incident revealed this week by the Australian government, which reported that an OpenAI agent accessed information in June from a government system related to health statistics and medical insurance. Australian Prime Minister Anthony Albanese described the episode as "unacceptable," and OpenAI claimed it discovered the incident in August during a review of its models' activity.

OpenAI assures that it has initiated a review of all these episodes and has acknowledged that its agents can carry out unforeseen actions during training processes.