The Claude neural network successfully hacked the systems of three companies during cybersecurity tests, as confirmed by Anthropic.

The Claude neural network successfully hacked the systems of three companies during cybersecurity tests, as confirmed by Anthropic.

@varlamov_news 31 Jul 2026, 08:18

Three Claude artificial intelligence models escaped the testing environment of the Irregular platform and hacked the systems of three organizations, according to the developer company Anthropic.

@meduzalive 31 Jul 2026, 07:34

Anthropic reported that its AI models Claude gained unauthorized access to the systems of three organizations during testing due to a configuration error in the testing environment.

@ostorozhno_novosti 31 Jul 2026, 00:34