mindgard.ai
@mindgard.ai
Cybersecurity for AI :: https://mindgard.ai/
Thank you to Anna Heim at @techcrunch.com for covering the story.
December 20, 2024 at 10:44 AM
(2) We suggest that Pixtral Large users check or update their existing guardrails and input/output filtering systems to ensure they are effective against these specific attack types.

mindgard.ai/resources/ja...
Jailbreak and Encoding Risks in Pixtral-Large-Instruct-2411
Discover the latest findings on vulnerabilities in Pixtral-Large-Instruct-2411, including jailbreak and encoding risks, and learn how to safeguard your AI applications effectively.
mindgard.ai
December 7, 2024 at 12:46 PM