AI security insights

Each insight solves one engineering or governance problem in depth, with the evidence behind it.

Enterprise AI defence

Mitigating Indirect Prompt Injection in Multi-Tenant RAG Pipelines
How untrusted documents, emails and web pages hijack RAG sessions, and the privilege boundaries and deterministic output checks that contain the damage.
Securing Autonomous AI Agents: Deterministic Gateways for Tool Use and Function Calling
AI agent security fails when agents hold their own credentials. A policy gateway, strict tool-call schemas and rare, clear approvals contain excessive agency.