Check Point researchers presented findings at Black Hat about vulnerabilities in AI agent frameworks used by enterprises to build AI applications. Their core argument is that prompt injection is a symptom, not the root cause — the real security problem lies in the design of the AI agent frameworks themselves.
Table of contents
This one time, at Hacker Summer Camp …Google dev kit spurs first-ever agent-on-agent violenceAI researchers let models off the leash – then watched as they tried to add malware to a FOSS projectBypassing AI guardrails is so easy a script kiddie can do it1 Impression