Academy module 2 · ISO/IEC 27090 · Lesson 2 of 8
Understand poisoning and backdoor risk
About 8 minutes · Introductory draft · Checked 2026-10-23
AI-authored and AI editorially reviewed; independent review pending. Read course scope and remaining gaps.
Session only. Enable saving or download a backup from this course’s hub.
Completion is self-reported and does not certify understanding.
See it · Try it · Recall it
Picture the topic
Remember this: Trust the training process, not just the file checksum.
Explore the anchors
Understand where data and labels came from.
Concept matching activity
Match this task or description to its concept:
Look for harmful patterns and anomalous influence.
Choose a concept before checking the explanation.
A quick way to study this topic
- Hide the explanation and redraw these three anchors from memory.
- Explain the difference between the anchors in your own words.
- Apply the distinction to the worked example below. Check the full lesson for assumptions and exceptions.
- Recall it again tomorrow, then after a few days. A memory hint is a retrieval cue, not a replacement for understanding.
Original academy illustrations and fictional exercises. These are learning simulations, not a real virtual machine, network assessment or measure of job readiness. Activities reset when you leave or reload; they do not change quiz scores.
What you will learn
- Understand where data and labels came from.
- Look for harmful patterns and anomalous influence.
- Evaluate behavior before approving a new model.
Concepts and distinctions
Poisoning involves adversarial influence over training data or the learning process. An attacker may seek broad degradation or a targeted behavior that appears only under particular conditions. A backdoor can be difficult to notice when ordinary test results remain strong. Benign collection errors can also damage a model, so the investigation should distinguish evidence of manipulation from accidental quality failures.
A checksum can reveal whether an artifact differs from a trusted reference. It cannot establish that the reference was trustworthy or that the dataset’s labels are appropriate. Data provenance, access control, review, separation of responsibilities and evaluations with suitable coverage work together. A signed but maliciously curated dataset can still produce harmful learned behavior.
Apply the process and collect evidence
Control who can add or relabel training examples and how changes are approved. Preserve dataset versions and transformation records so investigators can reproduce a release. Evaluate relevant slices and targeted conditions instead of looking only at aggregate accuracy. Stop or quarantine a suspect update under an approved process while evidence is assessed; do not experiment on production learning data without authority.
Work through a scenario
Fictional facts for learning. This exercise is not scored.
A retraining batch has a valid hash but an unexpected concentration of mislabeled examples associated with one customer segment. Overall accuracy changes very little.
Compare your reasoning
The hash supports file identity only relative to its reference. Investigate label provenance, who supplied the examples, affected segments and behavior under relevant conditions. Preserve the release inputs and compare trusted evaluations before approving deployment.
Remember
- Trust the training process, not just the file checksum.
- Understand where data and labels came from.
- Evaluate behavior before approving a new model.
Orientation and further reading
- ISO: ISO/IEC 27090:2026 · Published AI-security guidance
- NIST AI 100-2 E2025: adversarial machine learning
Original examples and explanations. Background references do not establish complete exam coverage.
Try ISO/IEC 27090 practice questions · Report a content correction