Draft high-risk AI serious-incident guidance

Official metadata

Official title

DRAFT GUIDANCE ARTICLE 73 AI ACT- INCIDENT REPORTING (HIGH-RISK AI SYSTEMS)

Document type

Guidelines

Document date

2025-09-26

Document date kind

Publication

Document-date evidence

European Commission · Commission consultation page, Publication field (26 September 2025) and Downloads entry 'Draft Guidance article 73 AI Act - incident reporting'. — Establishes 2025-09-26 as this exact record's publication-based document date.

Publication date

2025-09-26

Publication-date evidence

European Commission · Commission consultation page, Publication field (26 September 2025) and Downloads entry 'Draft Guidance article 73 AI Act - incident reporting'. — Establishes the dated official consultation-page publication associated with this download; it does not independently establish the earliest release of every binary revision.

Version label

Consultation draft

Legal status

Non binding

Language

en

Research classifications

These are researcher classifications used to organise and analyse the corpus; they are not official EU metadata.

Record level

Principal

Version status

Draft

Temporal collection

Contemporary EU AI policy

Relevance class

Direct AI relevance

Policy stage

Consultation

Corpus tier

Core

Sectors

• General cross sector

Policy membership

• AI Act implementation
• EU AI policy pathway

Classification evidence

• Relevance class: Direct AI relevance · European Commission · PDF page 1 heading 'DRAFT GUIDANCE ARTICLE 73 AI ACT- INCIDENT REPORTING (HIGH-RISK AI SYSTEMS)' and scope paragraphs on Article 73 reporting for high-risk AI systems, including the express exclusion of systemic-risk GPAI reporting. — The artifact directly governs, interprets, evaluates or operationalises EU artificial-intelligence policy rather than merely mentioning AI.
• Sector tags: General cross sector · European Commission · PDF page 1 heading 'DRAFT GUIDANCE ARTICLE 73 AI ACT- INCIDENT REPORTING (HIGH-RISK AI SYSTEMS)' and scope paragraphs on Article 73 reporting for high-risk AI systems, including the express exclusion of systemic-risk GPAI reporting. The cited scope applies horizontally to AI systems, models, providers, deployers or affected persons rather than one economic sector. — The retained general_cross_sector tag is supported by the artifact's horizontal AI-policy scope.
• Provenance tags: Authored by an EU institution · European Commission · Consultation page field 'Publication 26 September 2025' and PDF 119624 cover 'DRAFT GUIDANCE ARTICLE 73 AI ACT- INCIDENT REPORTING (HIGH-RISK AI SYSTEMS)'. — The exact institutional artifact supports the retained EU-institution provenance classification.
• Provenance tags: Official consultation material · European Commission · Consultation page field 'Publication 26 September 2025' and PDF 119624 cover 'DRAFT GUIDANCE ARTICLE 73 AI ACT- INCIDENT REPORTING (HIGH-RISK AI SYSTEMS)'. The official page and exact file label the material as a draft consultation instrument. — The exact artifact is expressly a consultation draft.
• Provenance tags: Officially published · European Commission · Official consultation page metadata field 'Publication 26 September 2025' beside 'Draft Guidance' download 119624. — The exact version is distributed through an official EU institutional endpoint.

Production provenance

Provenance tags

• Authored by an EU institution
• Official consultation material
• Officially published

Named bibliographic authors

None recorded after review

Institutions and roles

• European Commission · Author · role evidence · Consultation page field 'Publication 26 September 2025' and PDF 119624 cover 'DRAFT GUIDANCE ARTICLE 73 AI ACT- INCIDENT REPORTING (HIGH-RISK AI SYSTEMS)'.

Official sources and identifiers

• Official source · Publisher/host: European Commission
• Official source · Publisher/host: European Commission

Research assessment

Inclusion rationale

Verified official English text within the AI Act implementation and amendment corpus.

Researcher notes

Official metadata is transcribed from the cited source; corpus tier and policy-stage classification are researcher assessments. Subsequent file-level review by AI-assisted reviewer at 2026-09-05T08:20:53Z: retrieved official PDF 119624 and recorded its real SHA-256 and local file-write-completion timestamp (an operational retrieval-time proxy). Complete selected pages transcribed for identity/scope review; pages 4-12 not reviewed. Page 1 identifies Article 73 high-risk incident reporting and expressly excludes systemic-risk GPAI reporting. The review copy is local only, not a committed archive. This narrow review does not reverify the other metadata or classifications. Identity review by AI-assisted reviewer at 2026-09-05T08:35:01Z: The separately titled Article 73 guidance is a substantive standalone policy text with its own background, objectives, definitions and analysis. Its principal classification describes standalone documentary identity, not final adoption. The title is transcribed from official PDF 119624, page 1, with line breaks joined. Draft status remains explicit; the other incident document is not its parent or another version. This field-limited review preserves all other metadata, classifications and earlier review attribution.

Review status

verified

Relationships

No additional published relationships touch this record.

Verification

Expanded evidence review

Verified

Reviewed by

Yichen Hao

Evidence review date

2026-09-05T18:07:46Z

The evidence review date is not a separate human sign-off date. Review credit and evidence-review actions

• Last verified: 2026-09-05T17:58:09Z · AI-assisted reviewer retrieved this official download with normal TLS verification. The 341927-byte response is a 13-page PDF, confirmed by PDF signature, Poppler and pypdf despite the malformed HTTP Content-Type value '/'. SHA-256: f1d58ecf69004b361d2201ea44e6219f00c27ba249ad5011965152de9167686d. retrieved_at uses observed UTC local file-write completion immediately following download as an operational retrieval-time proxy; it does not claim precise HTTP timing and is not the publication date. Complete selected pages transcribed for identity/scope review; pages 4-12 not reviewed. Page 1 identifies Article 73 high-risk incident reporting and expressly excludes systemic-risk GPAI reporting. Original bytes exist only in a local ignored review directory; no committed archive is claimed. Exact observation limits and timestamps are recorded in research/migrations/2026-09-05-official-pdf-evidence.json. Live re-review by AI-assisted reviewer on 5 September 2026 checked the official endpoint and exact-version identity, substantive scope, relevant dates and institutional attribution; prior verification history and retrieved_at are retained.
• Last verified: 2026-09-05T17:58:09Z · Official consultation opened 26 September 2025 and supplies separate draft guidance and reporting-template files (newsroom documents 119624 and 119623). Follow-up by AI-assisted reviewer at 2026-09-05T08:20:53Z: actual linked PDF bytes were retrieved and narrowly inspected; individual source records, checksums, selected page locators and local-only retention are recorded in research/migrations/2026-09-05-official-pdf-evidence.json. This supplements the earlier landing-page review; it is not a full-field or all-page re-verification, and unresolved relationship holds remain. Live re-review by AI-assisted reviewer on 5 September 2026 checked the official endpoint and exact-version identity, substantive scope, relevant dates and institutional attribution; prior verification history and retrieved_at are retained.