Cours · 5 chapitres
Adversarial AI
Defend and red-team your AI — prompt-injection foundations, defense-in-depth, attack taxonomy, automated jailbreak tooling, and context poisoning for production systems
Ce que tu sauras faire
- An orientation across the four-chapter Adversarial AI skill path — defender → threat-surface → offense → program — covering everything from your first prompt-injection patch to running an internal red-team practice that survives 2026 attacks.
- Why prompt injection exists, how attackers exploit it, and the layered defense every LLM-powered feature needs before shipping.
- Map and defend the agent-era attack surface — RAG poisoning, document-borne payloads, memory poisoning, and tool-output hijacking that direct-injection defenses don't reach.
- Attacker LRMs, BYO-attacker patterns, the DeepTeam / PyRIT / Mindgard trio, and the automated algorithms (PAIR, TAP, GCG) that turn red-teaming from a one-engineer pursuit into a continuous program.
- Run a red-team program for production AI — taxonomy, the finding lifecycle, runtime monitoring, regression discipline, and the internal playbook.
Ce qu’il contient
- 1Adversarial AI: Start Here
An orientation across the four-chapter Adversarial AI skill path — defender → threat-surface → offense → program — covering everything from your first prompt-injection patch to running an internal red-team practice that survives 2026 attacks.
- 2Prompt Injection Defense Foundations
Why prompt injection exists, how attackers exploit it, and the layered defense every LLM-powered feature needs before shipping.
- 3Context Poisoning & Indirect Injection
Map and defend the agent-era attack surface — RAG poisoning, document-borne payloads, memory poisoning, and tool-output hijacking that direct-injection defenses don't reach.
- 4Automated Jailbreak Tooling
Attacker LRMs, BYO-attacker patterns, the DeepTeam / PyRIT / Mindgard trio, and the automated algorithms (PAIR, TAP, GCG) that turn red-teaming from a one-engineer pursuit into a continuous program.
- 5AI Red Teaming & Adversarial Evaluation
Run a red-team program for production AI — taxonomy, the finding lifecycle, runtime monitoring, regression discipline, and the internal playbook.
Obtiens un certificat
Termine tous les chapitres pour recevoir ton certificat de réussite.