AI models are getting so good at finding vulnerabilities that some experts say the tech industry might need to rethink how ...

The once-distant prospect of AI models executing cyberattacks fully on their own now looks unavoidable, according to a range of recent academic studies and industry warnings. Why it matters: This is ...

Artificial intelligence is becoming smarter and more powerful every day. But sometimes, instead of solving problems properly, AI models find shortcuts to succeed. This behavior is called reward ...

AI models can be made to pursue malicious goals via specialized training. Teaching AI models about reward hacking can lead to other bad actions. A deeper problem may be the issue of AI personas. Code ...

This is the online edition of The Wiretap newsletter, your weekly digest of cybersecurity, internet privacy and surveillance news. To get it in your inbox, subscribe here. Tenzai's five cofounders are ...

AI security risks are shifting from models to workflows after malicious extensions stole chat data from 900,000 users & ...

WASHINGTON (AP) — A team of researchers has uncovered what they say is the first reported use of artificial intelligence to direct a hacking campaign in a largely automated fashion. The AI company ...

Researchers from MIT, Northeastern University, and Meta recently released a paper suggesting that large language models (LLMs) similar to those that power ChatGPT may sometimes prioritize sentence ...