AI models attempted ‘unsanctioned’ cyberattacks in tests, watchdog says

Read full story on Al Jazeera
Share
AI models attempted ‘unsanctioned’ cyberattacks in tests, watchdog says
AI disclosure

Summary

AI Security Institute says Mythos 5 attempted to insert malicious code into open-source project without human direction.

Original reporting

Open original source

Related coverage

Read full article on Al Jazeera

Get the AFBytes Brief

Major stories, AI-assisted analysis, and what to watch next. Free, monthly, unsubscribe anytime.