Anthropic's artificial intelligence models "gained unauthorized access" to three outside organizations during testing meant to keep them away from "real-world" systems, the company said on Thursday. Anthropic examined more than 141,000 "evaluation runs" and found that three different versions of its model, known as Claude, improperly accessed the systems of three unidentified organizations. Unlike a recent incident involving OpenAI's technology, Anthropic's models had internet access "due to a misunderstanding between us and our evaluation partner," called Irregular, Anthropic said in a blog post. Claude used "basic techniques, such as exploiting weak passwords and unauthenticated endpoints." The older model "continued its attack even after getting evidence it was running on the open internet" whereas the latest model stopped. "In none of these situations did Claude exfiltrate itself or deliberately attempt to escape its test environment," the firm said. The models included one of its most powerful, known as Mythos 5, released only to a limited number of approved partners. Anthropic is working with Irregular to assess the situation and has contacted or attempted to contact all three impacted organizations. OpenAI admitted last week its models broke out of their confined environment, connected to the internet, and infiltrated Hugging Face, later finding three additional incidents. The incidents triggered a petition signed by more than 1,000 employees at cutting-edge AI companies, including Anthropic CEO Dario Amodei, calling on the US government to support an international effort