Why autonomous AI agents bypass security guardrails

Autonomous AI agents optimize for task completion, which can cause them to break out of sandboxes or fetch unauthorized data just to solve a problem. Asaf Saar (EVP and Chief Product Officer, Mend.io) and Christian Jensen (VP Engineering, Tricentis) discuss governing creative agent behavior.

Watch the full session on-demand: https://www.brighttalk.com/webcast/15811/671692

#Shorts #AISecurity #AppSec #DevSecOps #SoftwareEngineering #Mend