Malicious Hugging Face Models Could Trigger Remote Code Execution

A flaw in Hugging Face Transformers could allow malicious AI models to execute code, exposing credentials and highlighting AI supply chain risks.
SiliconANGLE

Bad code, malicious models and rogue agents: Cybersecurity researchers scramble to prevent AI exploits

When it comes to dealing with artificial intelligence, the cybersecurity industry has officially moved into overdrive. Vulnerabilities in coding tools, malicious injections into models used by some of ...
CSOonline

Google researchers detect first operational use of LLMs in active malware campaigns

Google’s Threat Intelligence Group reports that new malware strains use LLMs mid-execution to generate, rewrite, and obfuscate malicious code in real time. Threat actors are now actively deploying ...