Notifications
Clear all
Topic starter
20/08/2026 5:43 pm
TL;DR: Counterfactual explanations and attribution-based explanations answer different governance questions in AI systems, with the former focusing on recourse and the latter on feature contribution, while both can be distorted by baseline choice, feature interactions, and feasibility constraints, according to Fiddler’s analysis. The practical lesson is that explainability only supports trust when teams treat it as a control for model governance, not a substitute for validation, monitoring, or accountability.
NHIMG editorial — based on content published by Fiddler: Understanding Counterfactual and Attribution Explanations in AI
Questions worth separating out
Q: How should teams decide between counterfactual and attribution explanations?
A: Use counterfactual explanations when the business need is recourse, appeal handling, or showing what would need to change for a different outcome.
Q: Why do attribution explanations change when the baseline changes?
A: Attribution methods compare the observed input to a reference point, so the baseline is part of the explanation itself, not just an implementation detail.
Q: What breaks when counterfactual explanations ignore feasibility?
A: They can recommend changes that are mathematically valid but impossible, irrelevant, or policy-violating in the real world.
Practitioner guidance
- Define the explanation purpose before deployment Separate user recourse, investigator review, and regulator-facing documentation into different explanation requirements so the same output is not forced to do all three jobs.
- Test counterfactual feasibility against policy Validate that any suggested change is operationally possible, legally acceptable, and consistent with the underlying causal structure of the data before exposing it to users.
- Stabilise attribution baselines Document the baseline dataset or reference point used for attribution and test whether the ranking of important features changes under alternative baselines.
What's in the full article
Fiddler's full blog post covers the detailed mechanics this analysis intentionally leaves for the source:
- Worked examples of counterfactual and attribution outputs across specific model types and feature sets
- Detailed discussion of optimisation, baselines, and sensitivity issues that shape explanation quality
- The original comparisons the authors use to show why the two explanation styles diverge in practice
👉 Read Fiddler's analysis of counterfactual and attribution explanations in AI →
Counterfactual vs attribution explanations: what do practitioners need to know?
Explore further
20/08/2026 6:26 pm
Explainability is a governance control, not a trust guarantee. Counterfactual and attribution methods improve visibility, but neither can certify that a model decision is fair, stable, or lawful. That matters for identity-adjacent AI use cases where the decision itself can affect access, onboarding, or fraud outcomes. Practitioners should treat explanation outputs as evidence to review, not as evidence that the model is safe to deploy.
A question worth separating out:
Q: How do security and risk teams operationalise AI explanations in governance?
A: Treat explanations as a control input, not a control outcome. Counterfactuals can support appeals and remediation, while attributions can support monitoring and model review. Teams should require reproducibility, baseline documentation, and a clear mapping between explanation type and decision workflow before relying on outputs in production.
👉 Read our full editorial: Counterfactual and attribution explanations sharpen AI model governance
