Unbounded AI use can break your systems

In this Help Net Security video, James Wickett, CEO of DryRun Security, explains cyber risks many teams underestimate as they add AI to products. He focuses on how fast LLM features are pushed into live applications without limits or guardrails.

The video also looks at AI-generated code and why it should not be trusted by default, especially for business logic and access control. Wickett closes by warning about blurred trust boundaries when AI systems can read code, call tools, and trigger workflows, turning small permission mistakes into major security problems.