---
id: CP-535
slug: ai-act-ai-literacy-article-4
lang: en
title: "AI literacy: what Article 4 of the AI Act requires"
description: What Article 4 of the AI Act asks organisations to do about AI literacy, and what changed in the version amended in 2026.
canonical_url: https://startbrain.io/cyberpedia/en/articles/ai-act-ai-literacy-article-4/
markdown_url: https://startbrain.io/cyberpedia/en/articles/ai-act-ai-literacy-article-4.md
publisher: Brain (https://startbrain.io)
license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
cite_as: "Brain, \"AI literacy: what Article 4 of the AI Act requires\", Cyberpedia, 2026. https://startbrain.io/cyberpedia/en/articles/ai-act-ai-literacy-article-4/ (CC BY 4.0)"
wing: 4
theme: use_responsibly
reflex: air-045
schema: G3
updated: "2026-11-14"
translations:
  en: https://startbrain.io/cyberpedia/en/articles/ai-act-ai-literacy-article-4/
  fr: https://startbrain.io/cyberpedia/fr/articles/ai-act-litteratie-ia-article-4/
  de: https://startbrain.io/cyberpedia/de/articles/ai-act-ki-kompetenz-artikel-4/
  es: https://startbrain.io/cyberpedia/es/articles/ai-act-alfabetizacion-ia-articulo-4/
  it: https://startbrain.io/cyberpedia/it/articles/ai-act-alfabetizzazione-ia-articolo-4/
  nl: https://startbrain.io/cyberpedia/nl/articles/ai-act-ai-geletterdheid-artikel-4/
  pl: https://startbrain.io/cyberpedia/pl/articles/ai-act-kompetencje-ai-artykul-4/
  pt: https://startbrain.io/cyberpedia/pt/articles/ai-act-literacia-ia-artigo-4/
  ja: https://startbrain.io/cyberpedia/ja/articles/ai-act-ai-literacy-article-4/
sources:
  - "n": 1
    title: Règlement (UE) 2024/1689 du Parlement européen et du Conseil du 13 juin 2024 (règlement sur l’intelligence artificielle)
    publisher: Publications Office of the European Union
    url: https://publications.europa.eu/resource/cellar/dc8116a1-3fe6-11ef-865a-01aa75ed71a1.0009.03/DOC_1
    published: "2024-07-12"
    accessed: "2026-11-14"
  - "n": 2
    title: "AI Act Service Desk - Article 4: AI literacy"
    publisher: European Commission
    url: https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-4
    accessed: "2026-11-14"
  - "n": 3
    title: AI Literacy - Questions & Answers
    publisher: European Commission
    url: https://digital-strategy.ec.europa.eu/en/faqs/ai-literacy-questions-answers
    accessed: "2026-11-14"
  - "n": 4
    title: "Entrée en vigueur du règlement européen sur l’IA : les premières questions-réponses de la CNIL"
    publisher: CNIL
    url: https://www.cnil.fr/fr/entree-en-vigueur-du-reglement-europeen-sur-lia-les-premieres-questions-reponses-de-la-cnil
    published: "2024-07-12"
    accessed: "2026-11-14"
  - "n": 5
    title: AI Act
    publisher: European Commission
    url: https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai
    accessed: "2026-11-14"
  - "n": 6
    title: Navigating the AI Act
    publisher: European Commission
    url: https://digital-strategy.ec.europa.eu/en/faqs/navigating-ai-act
    accessed: "2026-11-14"
  - "n": 7
    title: Repository of AI literacy practices
    publisher: European Commission
    url: https://digital-strategy.ec.europa.eu/en/policies/repository-ai-literacy-practices
    accessed: "2026-11-14"
  - "n": 8
    title: Living repository to foster learning and exchange on AI literacy
    publisher: European Commission
    url: https://digital-strategy.ec.europa.eu/en/library/living-repository-foster-learning-and-exchange-ai-literacy
    published: "2025-02-04"
    accessed: "2026-11-14"
---

# AI literacy: what Article 4 of the AI Act requires

Article 4 of the AI Act, the European regulation on AI, applies to providers and deployers of AI systems. It asks them to act on the AI literacy of their staff and of the other people who deal with the operation and use of these systems on their behalf [1]. The text published in 2024 speaks of ensuring, to their best extent, a sufficient level of AI literacy [1].

That text was amended in 2026 [4][5]. In the amended version, the obligation to take measures remains, but it does not require guaranteeing a specific level of AI literacy for any individual [2][3]. This page explains what the regulation, the European Commission and the CNIL say. It is not legal advice and does not settle any particular case.

## In 30 seconds

- Where it happens: in organisations that provide or deploy an AI system, for their staff and for the other people who operate or use it on their behalf [1][3].
- What the amended text asks: measures to support the development of AI literacy, taking into account each person's technical knowledge, experience, education and training, and the context of use [2].
- The reflex: before using an AI tool at work, be informed of the risks specific to that tool (hallucination, for example, for a tool such as ChatGPT) [3].
- On the organisation's side: an organisation can keep an internal record of trainings or other guiding initiatives, and the Commission says no certificate is needed [3].

## Straight answers

### What does "AI literacy" mean?

The 2024 regulation defines it as the skills, knowledge and understanding that allow providers, deployers and affected persons to make an informed deployment of AI systems. They also allow them to become aware of the opportunities and risks of AI and of the possible harm it can cause [1].

### Who is concerned?

Article 4 targets providers and deployers of AI systems [1]. A deployer is a person or body that uses an AI system under its own authority, except where the system is used in the course of a personal non-professional activity [1].

The European Commission specifies that the obligation concerns anyone in the organisation who deals directly with an AI system. The "other persons" are not employees but people broadly under the organisation's remit, for example a contractor, a service provider or a client [3]. According to the Commission, the AI Act's legal framework also applies outside the Union, as long as the AI system is placed on the Union market, used in the Union, or its use has an impact on people located in the Union [3].

### Is the 2024 text the one that applies today?

No. According to the CNIL, an amended version of the regulation was adopted on 24 July 2026 and entered into force on 27 July 2026 [4]. The European Commission also says that the Digital Omnibus on AI entered into force on 27 July 2026 [5].

The Commission's service desk page on Article 4 carries the amended version: it is based on the EUR-Lex consolidated version as at 27 July 2026, and the three paragraphs of Article 4 are marked as amended there [2]. The 2024 Official Journal [1] carries the original version. The Commission's FAQ on AI literacy describes the changes [3].

The differences, as the texts give them:

- 2024 text: take measures to ensure, to their best extent, a sufficient level of AI literacy [1].
- Amended version: take measures to support the development of AI literacy. The obligation does not require guaranteeing a specific level of AI literacy of any individual [2].
- The Commission and the Member States support and facilitate the efforts of providers and deployers, in particular SMEs. The Commission shall publish practical examples of how to comply on the single information platform [2].
- The AI Board shall adopt recommendations, taking into account European competence frameworks, to support the Commission and the Member States [2].

The Commission adds that, for organisations that deploy high-risk AI systems, the obligation to ensure that staff are trained to ensure human oversight remains in place [3].

### Since when?

Article 4 entered into application on 2 February 2025 [3][5][6]. The supervision and enforcement rules apply from 2 August 2026 [3]. This supervision is the task of the national market surveillance authorities, not of the AI Office [3].

### Is a mandatory training, a certificate or an AI officer required?

According to the Commission, no strict requirements or mandatory trainings are imposed, and there is no one-size-fits-all model [3]. Article 4 does not entail an obligation to measure employees' knowledge of AI [3]. No specific governance structure is mandated either [3]. However, simply relying on the instructions for use, or asking staff to read them, might be ineffective in many cases [3].

### What does an organisation risk if it does nothing?

According to the Commission, national market surveillance authorities could impose penalties and other enforcement measures, based on national laws. Any sanction must be proportionate, based on the individual case, and take into account factors such as the nature and gravity of the infringement and its intentional or negligent character. The Commission adds that this might be more likely if there is proof of an incident due to lack of appropriate training and guidance of employees or other persons [3].

## How it works

The Commission suggests steps to consider at a minimum when building a programme [3]:

- understand AI in general: what it is, how it works, what AI is used in the organisation, and what its opportunities and dangers are;
- consider the organisation's role: does it develop AI systems or does it just use systems developed by another organisation;
- consider the risk of the AI systems provided or deployed: what staff need to know and which risks they need to be aware of;
- build the AI literacy actions on that analysis, taking into account differences in knowledge, experience, education and training, as well as the sector and the purpose of use.

The Commission adds that these steps include legal and ethical aspects, and that connections to the EU AI regulation and to principles of ethics and governance are encouraged [3]. It does not impose requirements for specific sectors, but the sector and the purpose matter when designing the action [3].

The level of risk also plays a part. For a high-risk system, the Commission says that additional measures might be relevant so that staff know how to deal with it and how to avoid or mitigate its risks [3]. Among the examples of high-risk systems, the CNIL cites systems used in recruitment [4].

## What you do

The Commission sets no single method. Its answers do allow a four-step approach to be drawn out, for the organisation as well as for the employee.

- Map the uses. List the AI tools in use and ask whether they are developed in-house or provided by another organisation [3].
- Name the risks tool by tool. For an employee who uses ChatGPT to write advertising text or to translate, the Commission says they should be informed of the risks specific to the tool, for example hallucination [3].
- Adapt the training to the job and the level. Several levels of training or learning approaches can be appropriate, depending on the systems and on each person's knowledge [3]. For its own staff, the Commission describes learning packages aimed at three groups: generalists, managers and developers [3].
- Keep a record. An organisation can keep an internal record of trainings or other guiding initiatives [3].

SMEs can turn to the European Digital Innovation Hubs, now called Experience Centres for AI, which can serve as a first help desk on the AI Act [3].

## Pitfalls

- Treating the repository of practices as proof of compliance. The Commission has launched a repository of more than 40 AI literacy initiatives, run by companies and the public sector, to encourage exchange [7]. It writes that replicating these practices does not automatically grant a presumption of compliance with Article 4, and that it will use the repository to publish examples of compliance on the single information platform [7][8].
- Relying on the instructions for use alone. They might not be enough [3].
- Believing that a degree in AI removes any need for action. According to the Commission, a person with training or experience in AI development is normally considered AI literate, but this depends on the tool in question and their specific qualification. The organisation should still examine what this person needs to know, including legal and ethical aspects [3].
- Forgetting people who work for the organisation without being employees. According to the Commission, this depends on the concrete type of AI system and on the risk (for example for a high-risk system). In general, people working for a service provider or contractor need the appropriate AI skills for the task in question, like employees [3].
- Mixing up the two versions. The sufficient level in the 2024 text [1] no longer appears in the amended Article 4 [2].

## Sources

1. Publications Office of the European Union, "Règlement (UE) 2024/1689 du Parlement européen et du Conseil du 13 juin 2024 (règlement sur l’intelligence artificielle)", 12 July 2024. https://publications.europa.eu/resource/cellar/dc8116a1-3fe6-11ef-865a-01aa75ed71a1.0009.03/DOC_1 (accessed 14 November 2026)
2. European Commission, "AI Act Service Desk - Article 4: AI literacy". https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-4 (accessed 14 November 2026)
3. European Commission, "AI Literacy - Questions & Answers". https://digital-strategy.ec.europa.eu/en/faqs/ai-literacy-questions-answers (accessed 14 November 2026)
4. CNIL, "Entrée en vigueur du règlement européen sur l’IA : les premières questions-réponses de la CNIL", 12 July 2024. https://www.cnil.fr/fr/entree-en-vigueur-du-reglement-europeen-sur-lia-les-premieres-questions-reponses-de-la-cnil (accessed 14 November 2026)
5. European Commission, "AI Act". https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai (accessed 14 November 2026)
6. European Commission, "Navigating the AI Act". https://digital-strategy.ec.europa.eu/en/faqs/navigating-ai-act (accessed 14 November 2026)
7. European Commission, "Repository of AI literacy practices". https://digital-strategy.ec.europa.eu/en/policies/repository-ai-literacy-practices (accessed 14 November 2026)
8. European Commission, "Living repository to foster learning and exchange on AI literacy", 4 February 2025. https://digital-strategy.ec.europa.eu/en/library/living-repository-foster-learning-and-exchange-ai-literacy (accessed 14 November 2026)

---

Original source: Cyberpedia, the public cybersecurity and AI encyclopedia published by Brain. Canonical version: https://startbrain.io/cyberpedia/en/articles/ai-act-ai-literacy-article-4/
You may reuse this text freely under the CC BY 4.0 license (https://creativecommons.org/licenses/by/4.0/), provided you credit "Brain, Cyberpedia" with a link to the canonical URL above.
