Engineering / research
Secure the complete AI system
Analyze trust boundaries, identity, provenance, privacy, adversarial inputs, and evidence for security decisions.
Your route through engineering
Follow the path or begin with a question. Every lesson is open.
Your place, at your pace. Progress is optional and saved only in this browser when available.
Read / explore / reason
Inside this module
Module 01 research
Secure the complete AI system
Analyze trust boundaries, identity, provenance, privacy, adversarial inputs, and evidence for security decisions.
- 01
Threat modeling AI systems
Turn an AI architecture into explicit assets, adversary capabilities, trust boundaries, and testable security claims.
30 min - 02
Workload identity and least privilege
Connect short-lived service identities to resource authorization, credential lifetimes, and bounded operational authority.
29 min - 03
Model artifact provenance
Verify the identity and production history of model artifacts without confusing signatures with behavioral safety.
29 min - 04
Prompt injection and trust boundaries
Keep untrusted retrieved content from acquiring tool authority through an AI application's instruction channel.
32 min - 05
Training data poisoning and backdoors
Analyze data and model supply paths, targeted failures, and the limits of random audits when defending training pipelines.
31 min - 06
Privacy and membership inference
Evaluate what a model can reveal about training participation while accounting for base rates, access, and population mismatch.
31 min - 07
Differential privacy and accounting
Reason from neighboring datasets to sensitivity, randomized release, and cumulative privacy loss without confusing epsilon with an attack probability.
35 min - 08
Confidential computing and attestation
Understand protected execution, evidence freshness, and key release while keeping hardware trust separate from application correctness.
32 min - 09
Abuse detection and rate controls
Combine cost-aware admission, bounded bursts, and calibrated detection while respecting false positives and distributed enforcement limits.
31 min - 10
Security evaluation and response
Turn scoped assessments into reproducible evidence, release decisions, and practiced containment of AI-system incidents.
33 min