[tl;dr sec] #352 - Unprompted Slides, AWS AI Post-compromise Trends, Malicious LLM API Routers
// AI summary
Malicious LLM API routers are stealing credentials and injecting malicious code, alongside trends showing post-compromise activity involving patching binaries and finding cloud vulnerabilities.
// from the source
Patching vulnerabilities in binaries + finding cloud vulns, what different models do when given AWS creds, LLM routers stealing creds and injecting malicious code