Before you commission an AI pentest, know exactly what to test.
An AI pentest is only as good as its scope. The free checklist maps your AI estate against 190+ interactive checks — so your pentesting covers the risks that matter, with a board-ready scorecard to show for it.
Everything in the checklist
One working document that takes you from “what AI do we even have?” to a risk posture you can defend in front of the board.
The AI Risk Management Checklist
FREE · INTERACTIVEWhy we give this away
Mindgard grew out of a decade of AI security research at Lancaster University and runs the world’s largest AI security lab. The checklist is free because it’s the honest first step: once you know which controls you’re missing, the next question is whether the ones you have actually hold — and testing that is what Mindgard automates. No sales call is attached to the download.
Built from real findings, not best guesses
Before you download
Is this a compliance document or something my team will actually use?
Both, deliberately. The checks are practical actions — inventory, controls, safeguard testing, incident response, monitoring — and each one maps to NIST AI RMF, ISO/IEC 42001 and the EU AI Act, so the work you do doubles as your compliance evidence.
Who is it for?
CISOs, AI governance leads and ML engineering teams — anyone accountable for AI risk who needs a defensible, framework-mapped program rather than a blank page.
What happens when I submit the form?
You get the interactive checklist immediately. No call gets scheduled.
Know exactly where your AI risk stands
190+ checks, a board-ready maturity scorecard, and a NIST · ISO 42001 · EU AI Act crosswalk — free and yours in under a minute.
Want to see whether your safeguards actually hold? Book a live demo instead.