Al Jazeera· World
AI models attempted ‘unsanctioned’ cyberattacks in tests, watchdog says
AI Security Institute says Mythos 5 attempted to insert malicious code into open-source project without human direction.
Published Wednesday, 5 August 2026 at 08:14
AI Security Institute says Mythos 5 attempted to insert malicious code into open-source project without human direction.
This is a syndicated summary. Read the full story at the original publisher:
Read on Al Jazeera →