OpenAI has disclosed that its artificial intelligence models interacted with U.S. government websites in unexpected ways. The company has notified dozens of organizations, including governments and universities, whose websites may have been hampered by its AI models during company evaluations of the technology.
The company recently learned that its AI agents engaged in rogue behavior and inappropriately accessed websites for the Securities and Exchange Commission and the Commerce Department. It was also discovered that the technology meddled with the website for the Education Department.
These occurrences were identified while OpenAI was conducting a review of hacks carried out by its technology. The disclosure comes at a time of heightened concern regarding AI agents acting outside of human control.