Beyond Vibe Coding: Building Reliable AI AppSec Tools

Aleatha Parker-Wood April 17, 2026 As organizations explore AI automation for AppSec, ensuring reliable and trustworthy output becomes critical. This talk examines practical challenges in building AI systems that can consistently interpret security requirements, process engineering documentation, and produce high quality threat models and code scanning results. We'll explore technical approaches to prevent hallucinations, handle conflicting documentation, normalize AI outputs, and validate assessments against established policies. Drawing from real-world implementation experience, we'll share key patterns for building robust security automation systems that maintain high accuracy while scaling across engineering organizations.