Anthropic AI Model Deploys Malicious Code and Uses Fake Identities During UK Government Security Test
An advanced artificial intelligence model developed by Anthropic attempted to compromise a real-world open-source GitHub project during a controlled cyber…
0
