StandardGuidancev1.0.0

prEN 18229-2: AI trustworthiness framework part 2, accuracy and robustness

The draft European standard meant to make the appropriate level of accuracy and robustness under Article 15 measurable.

The official source remains authoritative. This is general information about obligations and not legal advice. See this object on the map

Address and citation

This object has an address of its own that is never renamed or reused. Store the identifier in your own file, not the title or the link.

Identifier: praxikon:eu:ai-act:standard:standard-pren-18229-2-accuracy-robustness
Payload hash (sha256): 0e893c917fe9435350acafd09346a9557045fb27dfdea30ab46091cf9cdb7e23

Citation line

Praxikon, "prEN 18229-2: AI trustworthiness framework part 2, accuracy and robustness", praxikon:eu:ai-act:standard:standard-pren-18229-2-accuracy-robustness@1.0.0, dataset praxikon:sys:registry:dataset:ai-act-implementation-graph 2.1.0 (schema 1.4.0), effective_at 2026-08-08T00:00:00.000Z, known_at 2026-08-14T00:00:00.000Z, sha256 0e893c917fe9435350acafd09346a9557045fb27dfdea30ab46091cf9cdb7e23

Version: 1.0.0
Legal time (effective_at): 8 August 2026
Knowledge time (known_at): 8 August 2026
Closed on: Not closed
Topics: standards

Review status: Placed against the official source (8 August 2026). Next check due by 4 February 2027. The check date is the knowledge date of this version; no later recheck has been recorded.

What this object links to

Every relation appears below as a path: from the source with its locator, through the conditions and exceptions of the object carrying the relation, to the consequence. A locator belongs to a statement in the data and not to a relation, so the source is the source anchor of the carrying object.

The obligation this hangs off

1 of 1 shown

The object belongs to this obligation. The source line it hangs off sits there.

Source
Official fact on this object, with its locator.
- CEN-CENELEC JTC 21: European standards under standardisation request M/613
  Locator: prEN 18229-2 (draft standard), CEN/CLC/JTC 21 under M/613
  praxikon:eu:ai-act:source:cen-cenelec-jtc21
  Open official source
- EU Artificial Intelligence Act 2024/1689
  Locator: Article 15; Article 40
  praxikon:eu:ai-act:source:reg-eu-2024-1689
  Open official source
Via
No condition or exception recorded on this object.
Consequence
- ObligationArticle 15: accuracy, robustness and cybersecurity
  praxikon:eu:ai-act:obligation:article-15-accuracy-robustness

What this object is about

1 of 1 shown

The object is about this role. Undifferentiated: it does not follow that the duty rests on this role.

Source
Official fact on this object, with its locator.
- CEN-CENELEC JTC 21: European standards under standardisation request M/613
  Locator: prEN 18229-2 (draft standard), CEN/CLC/JTC 21 under M/613
  praxikon:eu:ai-act:source:cen-cenelec-jtc21
  Open official source
- EU Artificial Intelligence Act 2024/1689
  Locator: Article 15; Article 40
  praxikon:eu:ai-act:source:reg-eu-2024-1689
  Open official source
Via
No condition or exception recorded on this object.
Consequence
- ActorProvider of an AI system
  praxikon:eu:ai-act:actor:provider

What this object states

Official fact

Attributable to a named primary source, with a locator. Where they differ, the official source prevails.

prEN 18229-2 (AI trustworthiness framework, Part 2: Accuracy and robustness) is the JTC 21 deliverable under M/613 for Article 15 of the AI Act, which requires high-risk AI systems to achieve an appropriate level of accuracy, robustness and cybersecurity and to declare accuracy metrics in the instructions for use. As at June 2026 the deliverable was at the drafting stage. It has not yet been published as an EN and is not cited in the Official Journal.
- Locator: prEN 18229-2 (draft standard), CEN/CLC/JTC 21 under M/613praxikon:eu:ai-act:source:cen-cenelec-jtc21Open official source
- Locator: Article 15; Article 40praxikon:eu:ai-act:source:reg-eu-2024-1689Open official source

Our interpretation

Our own reading. It can change without the law changing, and it is not the position of a supervisory authority.

Article 15 leaves open what "appropriate" means, and that open norm is precisely what this deliverable would fill in. Until it is finished, you must justify yourself which accuracy level fits your application and why. The common trap is putting a single percentage in the instructions for use without stating on which population and under which conditions it was measured: that is not a metric, it is a marketing figure. A supervisory authority will ask about the measurement setup.
- Locator: prEN 18229-2 (draft standard), CEN/CLC/JTC 21 under M/613praxikon:eu:ai-act:source:cen-cenelec-jtc21Open official source
- Locator: Article 15; Article 40praxikon:eu:ai-act:source:reg-eu-2024-1689Open official source

Recommended step

A practical step we consider appropriate. Not an obligation following from the Regulation.

Record a test protocol per high-risk system: which metric, which test population, which subgroups measured separately, and what performance degradation under out-of-distribution conditions is acceptable. Use the ISO/IEC 24029 series for robustness methodology and retain the raw test results, not only the summary.
- Locator: prEN 18229-2 (draft standard), CEN/CLC/JTC 21 under M/613praxikon:eu:ai-act:source:cen-cenelec-jtc21Open official source
- Locator: Article 15; Article 40praxikon:eu:ai-act:source:reg-eu-2024-1689Open official source

When this applies

No condition recorded on this object.

When this does not apply

No exception recorded on this object.

Referring to this object

Citation block

Copy this reference into your advice, article or file. The identifier, the version and the hash keep the statement findable later, even once the dataset has moved on.

Reference

Praxikon, "prEN 18229-2: AI trustworthiness framework part 2, accuracy and robustness",
praxikon:eu:ai-act:standard:standard-pren-18229-2-accuracy-robustness@1.0.0,
dataset praxikon:sys:registry:dataset:ai-act-implementation-graph 2.1.0 (schema 1.4.0),
effective_at 2026-08-08T00:00:00.000Z, known_at 2026-08-08T00:00:00.000Z,
sha256 0e893c917fe9435350acafd09346a9557045fb27dfdea30ab46091cf9cdb7e23,
https://www.praxikon.com/en/verkenner/standard/standard-pren-18229-2-accuracy-robustness
(https://www.praxikon.com/api/v1/entities?id=praxikon%3Aeu%3Aai-act%3Astandard%3Astandard-pren-18229-2-accuracy-robustness&effective_at=2026-08-08&known_at=2026-08-08&lang=en, accessed 2026-08-25)

Short form

praxikon:eu:ai-act:standard:standard-pren-18229-2-accuracy-robustness@1.0.0 (sha256 0e893c91)

BibTeX

@misc{praxikon-eu-ai-act-standard-standard-pren-18229-2-accuracy-robustness-1-0-0,
  author       = {{Praxikon}},
  title        = {prEN 18229-2: AI trustworthiness framework part 2, accuracy and robustness},
  year         = {2026},
  version      = {1.0.0},
  number       = {praxikon:eu:ai-act:standard:standard-pren-18229-2-accuracy-robustness},
  howpublished = {AI Act Change \& Evidence Graph, dataset 2.1.0, schema 1.4.0},
  note         = {effective_at 2026-08-08T00:00:00.000Z; known_at 2026-08-08T00:00:00.000Z; sha256 0e893c917fe9435350acafd09346a9557045fb27dfdea30ab46091cf9cdb7e23},
  url          = {https://www.praxikon.com/en/verkenner/standard/standard-pren-18229-2-accuracy-robustness},
  urldate      = {2026-08-25},
  language     = {en}
}

CSL JSON

[
  {
    "id": "praxikon:eu:ai-act:standard:standard-pren-18229-2-accuracy-robustness@1.0.0",
    "type": "dataset",
    "title": "prEN 18229-2: AI trustworthiness framework part 2, accuracy and robustness",
    "container-title": "AI Act Change & Evidence Graph",
    "publisher": "Praxikon",
    "version": "1.0.0",
    "number": "praxikon:eu:ai-act:standard:standard-pren-18229-2-accuracy-robustness",
    "URL": "https://www.praxikon.com/en/verkenner/standard/standard-pren-18229-2-accuracy-robustness",
    "language": "en",
    "issued": {
      "date-parts": [
        [
          2026,
          8,
          8
        ]
      ]
    },
    "accessed": {
      "date-parts": [
        [
          2026,
          8,
          25
        ]
      ]
    },
    "note": "dataset praxikon:sys:registry:dataset:ai-act-implementation-graph 2.1.0; schema 1.4.0; effective_at 2026-08-08T00:00:00.000Z; known_at 2026-08-08T00:00:00.000Z; sha256 0e893c917fe9435350acafd09346a9557045fb27dfdea30ab46091cf9cdb7e23; retrieved_from https://www.praxikon.com/api/v1/entities?id=praxikon%3Aeu%3Aai-act%3Astandard%3Astandard-pren-18229-2-accuracy-robustness&effective_at=2026-08-08&known_at=2026-08-08&lang=en; licence https://www.praxikon.com/nl/legal/terms"
  }
]

How to verify a reference later is set out in the methodology. Terms

For agents and integrations

This page and the machine output come from the same object and the same two time axes.

Open as JSON Open as JSON-LD Read the methodology