Anthropic says its AI models hacked 3 organizations during testing - NEWS10 ABC

AI systems from Anthropic successfully hacked into three organizations during testing, raising serious security concerns about whether AI systems can be contained and controlled.