How do I use the Instructions for Use Generator?

Identify provider and system details Document intended purpose, capabilities, and performance levels Specify accuracy metrics, known limitations, and failure modes Define human oversight measures and technical requirements Describe logging capabilities and retention periods Export and include with every deployment of the system

Omnibus update (7 May 2026): Annex III high-risk provisionally shifted from 2 Aug 2026 to 2 Dec 2027. Art. 50 transparency (chatbots, deepfakes, emotion, biometric) still 2 Aug 2026. Art. 4 / Art. 5 / GPAI live now. Details.

Why this matters

Art. 13(3) mandates that providers supply specific instructions with every high-risk AI system. Without this document, a system cannot legally be placed on the EU market. It is also required content for EU database registration (Annex VIII).

How to use it

1Identify provider and system details
2Document intended purpose, capabilities, and performance levels
3Specify accuracy metrics, known limitations, and failure modes
4Define human oversight measures and technical requirements
5Describe logging capabilities and retention periods
6Export and include with every deployment of the system

Legal basis

Art. 13
Art. 13(3)
Annex IV §1

Instructions for Use Generator

Mandatory under Art. 13(3) EU AI Act. Fill in the narrative fields plus the structured performance, misuse and input-data sections before exporting.

AI System Name

Version

Issue Date

0/8 narrative fields completed

Provider Identity & Contact

Art. 13(3)(a)

System Identification

Art. 13(3)(b)

Intended Purpose

Art. 13(3)(c)

Capabilities & Limitations

Art. 13(3)(d)

Human Oversight Measures

Art. 13(3)(e)

Technical Measures & Hardware

Art. 13(3)(f)

Logging & Audit Trail

Art. 13(3)(g)

Expected Lifetime & Updates

Art. 13(3)(h)

Performance Levels & Accuracy

Art. 13(3)(b)(ii) / Art. 13(3)(d)(i)

Each row captures one performance metric with its value, the test conditions, the demographic split and the confidence interval. Minimum one row required to export.

Metric #1

Metric name

Value

Test conditions (dataset, version, distribution, evaluator)

Demographic disaggregation

Confidence interval

Reasonably Foreseeable Misuse

Art. 13(3)(d)(ii)

Tick every category that applies. The "Other" row captures misuse that does not fit the standard taxonomy. At least one category must be selected to export.

Off-label use — Deployment outside the system’s intended purpose.Distribution shift — Input data drifts from the training distribution.Autonomy without oversight — Art. 14 human oversight bypassed in practice.Geographic / linguistic mismatch — System used in a region or language it was not trained on.Adversarial input — Prompt injection, evasion attack, or other adversarial input.User overestimating capability — Over-reliance on system output (automation bias).

Other (free text)

Input Data Specification

Art. 13(3)(d)(iii)

Specifications of the input data the system requires and what the system will reject.

Required fields

Format

Rejection criteria

Sanitisation

Max input size

Allowed encodings

Export blocked. 7 required fields missing: AI system name, Version, Provider identity, +4 more

Important Legal Disclaimer

This tool is a self-assessment aid only and does not constitute legal advice, a formally certified compliance assessment, or an independently audited report.

Outputs — including reports, scores, checklists, and generated documents — are for internal use and should be reviewed by a qualified legal representative or independent AI compliance auditor before being relied upon for regulatory, procurement, or public-disclosure purposes.

This tool does not replace a notified body conformity assessment where one is required under Art. 43(1) of the EU AI Act (e.g. biometric identification systems for law enforcement).

All assessment risk lies with the user. AIAuditRef, its developers, and staff accept no liability for losses arising from use of or reliance on these outputs. Always verify against official sources: the EU AI Act (Regulation 2024/1689) and your national enforcement authority.