EDITION 11 · EU AI ACT FOR BUSINESS

What does the EU AI Act's AI literacy obligation require from companies?

29 July 2026 · Matúš Paško

The Week in One Line

The EU AI Act's AI literacy requirement has applied since 2 February 2025. It covers most organisations that use AI, not just those using high-risk AI systems.

What the Act Actually Says

Article 4 of the EU AI Act requires providers (companies that develop or sell AI systems) and deployers (organisations that use AI systems) to make sure their employees and anyone operating AI systems on their behalf have an appropriate level of AI literacy.

The level of training should reflect the person's technical knowledge, their role, and how the AI system is being used.

Unlike many other parts of the AI Act, Article 4 isn't limited to high-risk AI systems.

It can apply to organisations using everyday AI tools such as:

Chatbots
Content generation tools
Scheduling assistants
AI meeting assistants
AI coding tools

This means most organisations using AI in the EU already have an obligation under the AI Act.

The AI Act defines AI literacy as the skills, knowledge, and understanding needed to make informed decisions about using AI and to understand both its benefits and its risks.

The obligation isn't limited to employees. It also covers people working with AI systems on behalf of your organisation, such as:

Contractors
Service providers
External teams operating AI systems for your business

The AI Act does not require organisations to:

Test or certify employees
Appoint a dedicated AI Officer
Create a formal AI governance board solely to comply with Article 4

One important point to understand:

The AI Act doesn't create a separate fine specifically for failing to meet Article 4. However, regulators may consider poor AI literacy when assessing broader AI Act compliance failures, particularly where a lack of training contributes to other breaches.

One Development to Watch

The Digital Omnibus proposal (the EU's proposal to simplify parts of the AI Act) includes a change to Article 4.

If adopted, organisations would no longer be required to ensure a sufficient level of AI literacy. Instead, they would be required to support the development of AI literacy among their staff.

This would give organisations greater flexibility while still encouraging responsible AI use.

For now, however, the current version of Article 4 remains in force.

Sources: artificialintelligenceact.eu – Article 4; European Commission AI Literacy Q&A (May 2025); Travers Smith; Ropes & Gray; Gibson Dunn.

What This Means for Your Business

This requirement may already apply to your organisation today, even if you don't use any high-risk AI systems.

Using AI tools such as ChatGPT, Microsoft Copilot, Gemini, or AI-powered scheduling assistants may be enough to trigger Article 4.

The European Commission has confirmed that supervision and enforcement of this requirement begins on 3 August 2026, separate from the delayed timeline for high-risk AI obligations.

The good news is that compliance doesn't require expensive certification programmes.

Instead, organisations should make sure employees understand how the AI tools they use work, what their limitations are, and what risks they introduce.

The training should also reflect each person's role.

For example:

Someone using AI to screen job applicants needs different training from someone using AI to summarise meeting notes.
Software engineers need different guidance than HR teams.
Customer-facing staff may need different training than compliance or legal teams.

This Week's Action Point

Ask yourself three simple questions:

Which AI tools are people using across the organisation?
Do employees understand how those tools work and where they can make mistakes?
Do we have any documented AI guidance or training in place?

If the answer is no, start with a simple inventory of the AI tools used by each team.

That inventory becomes the foundation for building an AI literacy programme that's appropriate for your organisation.