Model Card

Use this template to responsible AI documentation for [machine learning model].

Template Metadata

Model Overview

Name, version, owner, model type, release status, and summary.

Notes

[Add context, assumptions, exceptions, evidence links, screenshots, calculations, or reviewer comments.]

Intended Use

Approved use cases, users, inputs, outputs, and prohibited uses.

Notes

[Add context, assumptions, exceptions, evidence links, screenshots, calculations, or reviewer comments.]

Training Data

Describe source data, time period, features, labels, sampling, and exclusions.

Notes

[Add context, assumptions, exceptions, evidence links, screenshots, calculations, or reviewer comments.]

Evaluation

Report metrics by segment, benchmark comparisons, thresholds, and test dataset details.

Notes

[Add context, assumptions, exceptions, evidence links, screenshots, calculations, or reviewer comments.]

Limitations

Document known weaknesses, edge cases, drift risks, and confidence boundaries.

Notes

[Add context, assumptions, exceptions, evidence links, screenshots, calculations, or reviewer comments.]

Risks

Identify fairness, privacy, security, safety, and operational risks with mitigations.

Notes

[Add context, assumptions, exceptions, evidence links, screenshots, calculations, or reviewer comments.]

Monitoring

Define production metrics, alerts, retraining triggers, and review cadence. Use Markdown tables for metrics and risks.

Notes

[Add context, assumptions, exceptions, evidence links, screenshots, calculations, or reviewer comments.]

Review and Signoff

Document review conclusions, approvals, unresolved items, and next review date.