Black Hat USA 2026 – Summary of Vendor Announcements (Part 3)

Many companies are showcasing their products and services this week at the 2026 edition of the Black Hat conference in Las Vegas. The post Black Hat USA 2026 – Summary of Vendor Announcements (Part 3) appeared first on SecurityWeek .

According to the article, CrowdStrike announced "AI Unlocked: Agents of Chaos," a global AI red teaming competition with AWS that explicitly challenges participants to exploit rogue AI agents using prompt injection and related techniques to study emerging agentic AI security risks.[1] Other vendors at Black Hat 2026 are also emphasizing agentic incident response and AI-native cyber defense, reflecting growing attention to offensive testing of AI systems.[1][3] From a RealGround perspective, this highlights that prompt injection against autonomous or semi-autonomous AI agents is now a mainstream, actively explored threat vector rather than a hypothetical risk, and organizations deploying agents should treat it as a first-class security concern. Practical implications include the need to harden agent architectures against untrusted inputs, continuously red team AI agents for injection and jailbreak paths, and establish governance around how agents access tools and data.