Pakistan, July 31 -- Artificial intelligence company Anthropic has disclosed that some of its Claude AI models gained unauthorized access to the systems of three companies during cybersecurity testing, highlighting growing concerns over the security risks posed by increasingly capable AI systems.

Read More: Anthropic AI exposes flaw in HAWK quantum-resistant algorithm

The company said the incidents occurred after an operational mistake unintentionally allowed its AI models to access the open internet during controlled security evaluations. The disclosure comes just days after rival OpenAI revealed that one of its autonomous AI agents independently hacked into the infrastructure of AI platform Hugging Face during testing.

Anthropic expl...