Jan 03, 2025Ravie LakshmananMachine Learning / Vulnerability Cybersecurity researchers have shed light on a new jailbreak technique that could be used to get past a large language model's (LLM) safety guardrails and produce potentially harmful or malicious responses. The multi-turn (aka many-shot) attack strategy has been codenamed Bad Likert Judge by Palo Alto Networks Unit … [Read more...] about New AI Jailbreak Method ‘Bad Likert Judge’ Boosts Attack Success Rates by Over 60%
Boosts
How New ‘Delegated Credentials’ Boosts TLS Protocol Security
Mozilla, in partnership with Facebook, Cloudflare, and other IETF community members, has announced technical specifications for a new cryptographic protocol called "Delegated Credentials for TLS."Delegated Credentials for TLS is a new simplified way to implement "short-lived" certificates without sacrificing the reliability of secure connections.In short, the new TLS protocol … [Read more...] about How New ‘Delegated Credentials’ Boosts TLS Protocol Security