Aug 20, 2026 · 3 min read · Christchurch, NZ
From code-gen to pen testing: what happened when we pointed an LLM at our dev environment
An external audit gave us a relatively clean bill of health. Out of curiosity I pointed an uncensored model at our staging environment anyway, effectively saying "hack my own software." It found things we'd missed.
aisecurityllm
Read essay →