VIIIProgram

AI Security

Attacking and defending AI systems.

  • 801Course

    Prompt Injection

    The attack every AI system faces.

    4 lessons·Practitioner
    The attack
  • 802Course

    Data Poisoning

    The training-time attack.

    3 lessons·Practitioner
    The attack
  • 803Course

    Model Theft and Extraction

    Stealing a model through its API.

    3 lessons·Practitioner
    The attack
  • 804Course

    Jailbreaks

    Bypassing the guardrails.

    3 lessons·Practitioner
    The category
  • 805Course

    Security Testing for AI

    The QA discipline for AI security.

    3 lessons·Practitioner
    The practice
  • 806Course

    AI Red Teaming

    The offensive discipline.

    3 lessons·Practitioner
    The engagement
  • 807Course

    Secure Deployment

    Defensive engineering.

    3 lessons·Practitioner
    The practice
  • 808Course

    Adversarial Examples

    The classical ML attack, updated.

    3 lessons·Practitioner
    The attack
  • 809Course

    Model Privacy

    Protecting training data and users.

    3 lessons·Practitioner
    The techniques