Locate the failure layer
Map failures to the human experience, system prompt, model, or infrastructure.
AILI Certification 07 / 08
Certified AI Red Team Practitioner CAIRT
Learn to test conversational AI and agents in an authorized, structured way. Find weaknesses, capture reproducible evidence, and turn the results into safeguards that teams can act on.
3 focused daysLive, intensive instruction
2-week capstoneApply the learning to real work
Live + appliedFrameworks, labs, and feedback
AILI credentialEarned through completed work
What you will be ready to do
Build judgment.Show your work.
Challenge AI systems before real users, attackers, or failures do.
This program combines live learning with applied practice so you leave with a repeatable approach and evidence of what you can do.
Run structured attacks
Practice prompt injection, persona hijacking, context manipulation, data extraction, bias probing, and boundary pushing.
Write actionable findings
Document the attempt, result, severity, ownership, and recommended change.
Build a testing rhythm
Define scope, authorization, team roles, readiness, cadence, and governance handoff.
Inside the intensive
The capabilitiesbehind the credential.
Failure modes and threat modeling
Understand hallucination, bias, jailbreak, data exposure, unsafe agency, and where each failure begins.
Authorized attack patterns
Test conversational systems with repeatable techniques, clear scope, and controlled evidence collection.
Layer-by-layer red team lab
Challenge a hardened assistant, log attempts, rate severity, and compare patterns across findings.
From finding to governance
Brief technical teams and executives, assign ownership, validate fixes, and establish a testing cadence.
The two-week capstone
Turn learning intoproof of capability.
Plan and execute an authorized red team assessment, then deliver reproducible findings, severity ratings, remediation recommendations, and a governance briefing.
- A threat model and test plan
- A red team scorecard
- Reproducible findings with severity ratings
- A remediation and governance briefing
Who this is for
For people who need tolead the work.
Program detail informed by Noelle Russell's current course material. Explore the related Maven workshop ↗
Security professionals and AI engineers testing AI systemsYou want practical frameworks, credible work products, and the judgment to use them responsibly.
Risk, governance, and assurance teams responsible for safe adoptionYou want practical frameworks, credible work products, and the judgment to use them responsibly.
Solution architects and responsible AI practitioners improving controlsYou want practical frameworks, credible work products, and the judgment to use them responsibly.
Ready to make this your next credential?
Tell us what you want to be able to do, and we’ll help you choose the right certification and next cohort.