# Derivation — eu-2026-1744/art_001/pt_009/a

> COMPOSED TEXT - NOT AUTHENTIC. This is the EU AI Act (Regulation (EU) 2024/1689) with the Digital Omnibus on AI (Regulation (EU) 2026/1744, in force 27 July 2026) applied by a deterministic, gated parser. Only the Official Journal publications are authentic law; no official consolidated version existed when this was generated. Corrigenda R(01)-R(04) do not correct the English text (verified against CELLAR). Every fragment's derivation chain is available in the source graphs (vault: data/graph/). Not legal advice.

- Page: https://eu-ai-act.standards.riskmandate.ai/derivations/art_001__pt_009__a.html
- Structured: https://eu-ai-act.standards.riskmandate.ai/derivations/art_001__pt_009__a.llm.json
- Text version: 2026-07-27 (generated 2026-07-31)
- Site version: v0.1.4
- Provisions root hash: `77c4ca7c208d73adc24b9a41eb29724bee29a41df98b786b85c57852fa3ae6bd`
- Instruction id: `eu-2026-1744/art_001/pt_009/a`
- Operation: replace · level: paragraph
- In force: 2026-07-27
- Payload sha256: `0cf6c4974c6eab1b7cea8d823c085f5003133c68de66c8cb74818ca9204dae62`
- OJ Formex member: `L_202601744EN.000101.fmx.xml` (CELEX 32026R1744)

## Context (enacting terms)

> Article 10 is amended as follows:

## Instruction (quoted official text)

> paragraph 1 is replaced by the following:

## Quoted payload (official replacement/inserted text)

1.High-risk AI systems which make use of techniques involving the training of AI models with data shall be developed on the basis of training, validation and testing data sets that meet the quality criteria referred to in paragraphs 2, 3 and 4 of this Article and in Article 4a(1) whenever such data sets are used.;

## Affected provisions

### `eu-2024-1689/art_010/par_001` — amended

**Before (as published):**

1. High-risk AI systems which make use of techniques involving the training of AI models with data shall be developed on the basis of training, validation and testing data sets that meet the quality criteria referred to in paragraphs 2 to 5 whenever such data sets are used.

**After (composed current text):**

1. High-risk AI systems which make use of techniques involving the training of AI models with data shall be developed on the basis of training, validation and testing data sets that meet the quality criteria referred to in paragraphs 2, 3 and 4 of this Article and in Article 4a(1) whenever such data sets are used.

- sha256 of composed text: `f904468aa7b500f55b155b494cfe88a7ff9ebadc09ceeae22cbd9881ab42c44a`
- derivation.json: https://eu-ai-act.standards.riskmandate.ai/provisions/eu-2024-1689/art_010/par_001/derivation.json

## How to check this

Compare the quoted instruction and payload above (hash-anchored to the OJ Formex bytes of CELEX 32026R1744) against the before/after text. That is the whole verification for this provision — about two minutes. Report the result, including the root hash above: https://github.com/Risk-Mandate/Risk-Mandate__EU-AI-Act__Current-Text/issues/new?template=provision-check.yml