AI Safety & Red-Teaming
Test, harden, and govern LLM systems against misuse and failure — prompt injection, jailbreaks, agentic risks, and the defense-in-depth reality that there is no single fix.
As LLMs move into products and agents, they bring a new attack surface: prompt injection, jailbreaks, data poisoning, and the risks of giving models real-world tools. This course teaches ML engineers, security professionals, and AI product teams how to think about AI threats, red-team systems, and build layered defenses — grounded in the actual frameworks (OWASP, NIST, MITRE ATLAS) and research.
It is defensive and educational: it teaches the categories of attack to test and defend against, not step-by-step exploit recipes. Current for 2026. A recurring theme: prompt injection is an unsolved problem, so defense-in-depth — not a silver bullet — is the realistic posture.
What you'll be able to do
- Understand the LLM threat landscape (prompt injection, jailbreaks, agentic risks)
- Red-team AI systems using manual and automated methods
- Build layered, defense-in-depth mitigations
- Apply the governing frameworks — OWASP, NIST, MITRE ATLAS
Curriculum
Module 1: Foundations
FreeWhy AI safety matters, the threat landscape, what red-teaming is, and the defense-in-depth mindset.
Module 2: The Threat Landscape in Depth
PremiumPrompt injection, jailbreaking, data and privacy attacks, and agentic risks — defensively.
- Prompt injection: direct and indirect15 min
- Jailbreaking: why safety guardrails fail12 min
- Data and privacy attacks12 min
- Agentic risks: when models can act12 min
Module 3: Red-Teaming Practice
PremiumManual vs. automated red-teaming, the tools and frameworks, lab approaches, and safety evals.
- Manual vs. automated red-teaming12 min
- Red-teaming tools and frameworks12 min
- How frontier labs and institutes red-team12 min
- Safety evals and measuring robustness12 min
Module 4: Defenses & Governance
PremiumLayered defenses, alignment techniques, the governing standards, and where the field is heading.
- Layered defenses and guardrails15 min
- Alignment: RLHF and Constitutional AI12 min
- Governance: NIST, MITRE ATLAS, and the EU AI Act12 min
- The road ahead12 min
Get the full course — one-time $29.
Start Module 1 free today. Buy once for lifetime access to the remaining 3 modules — no subscription.
Want all 53? Get the All-Access Bundle for $99 — one purchase, every course.