EU AI Act: What Article 4 Requires - and How to Prove It

Q: How does the diagnostic relate to the EU AI Act?

Article 4 has required AI literacy from staff who use AI since 2 February 2025, and proving it is process-based rather than certified. Every run leaves a dated, role-aware record of who was assessed and how they scored, which is evidence for the role-based assessment part of that proof. It is one input to your compliance file, not a certificate and not compliance on its own.

Q: Does Article 4 of the EU AI Act apply to my company?

If any of your employees use AI tools at work - ChatGPT, Copilot, Gemini, or others - your company is a 'deployer' under the [EU AI Act](/en/ai-compliance/ai-act-eu) and Article 4 applies to you. This covers virtually every company using AI in the EU, regardless of size or industry.

Q: What does 'AI literacy' mean under the EU AI Act?

AI literacy is defined as the skills, knowledge, and understanding needed to make informed decisions about AI deployment and to be aware of AI's opportunities, risks, and potential harm. It's not about becoming an AI expert - it's about understanding what AI can do, what it can't, and how to use it responsibly in your specific role.

Q: Is there an official EU certification for AI training?

No. There is no central EU certification for AI literacy training. Compliance is process-based - you demonstrate it through training logs, syllabus mapping, and documented role-based assessments. ISO/IEC 42001 (AI Management Systems) serves as a gold-standard reference framework.

Q: What documentation proves compliance with Article 4?

Companies need three things: (1) Training logs showing who received training, when, and what content they completed. (2) Syllabus mapping proving training covers relevant risks and contexts. (3) Role-based assessments documenting that individual backgrounds were considered when assigning training levels.

Q: When did the AI literacy requirement become effective?

Article 4 has been enforceable since 2 February 2025. This is not a future deadline - it's already in effect. The full EU AI Act regulation applies from 2 August 2026, but the AI literacy requirement is part of Chapters I and II which applied first.

Q: What are the penalties for non-compliance with the EU AI Act?

The [EU AI Act](/en/ai-compliance/ai-act-eu) provides for penalties up to EUR 35 million or 7% of global annual turnover for the most serious violations. While Article 4 penalties may be at the lower end, demonstrating non-compliance weakens your organization's overall AI Act posture. Proportionality clauses apply for SMEs and startups.

Q: Can existing corporate training satisfy Article 4?

Potentially, but only if it meets Article 4's specific requirements: training must be role-appropriate (taking into account technical knowledge and experience), context-specific (covering the actual AI tools your organization uses), and documented (with training logs and completion records). Generic 'AI awareness' seminars typically don't satisfy these requirements.

Q: How does AITutoro help with EU AI Act compliance?

Article 4 proof is process-based, and one of the three things an auditor looks for is a role-based assessment showing you took individual backgrounds into account. The diagnostic produces exactly that: a dated record of who was assessed, in what role, with what result. It also tells you who needs training and on what, which is where the other two pieces - training logs and syllabus mapping - come from once training is delivered.

Article 4 mandates AI literacy for all staff using AI. Effective since 2 February 2025. Here is what you need to know - and why a measured baseline of your workforce is the first piece of proof.

What You Need to Know

The EU AI Act (Regulation (EU) 2024/1689) is the world's first comprehensive legal framework for artificial intelligence. Among its provisions, Article 4 requires all providers and deployers of AI systems to ensure their staff has a sufficient level of AI literacy.

This isn't optional, and it's not future-dated. The AI literacy requirement has been enforceable since 2 February 2025. Every company that uses AI tools in the EU - from ChatGPT to Microsoft Copilot - must ensure their people understand what they're working with.

The practical starting point: assess where every employee stands today. A dated, role-aware diagnostic record is the foundation the rest of your compliance file is built on.

Quick Reference

Regulation: Regulation (EU) 2024/1689
Key Article: Article 4 - AI Literacy
Effective: 2 February 2025
Full enforcement: 2 August 2026

Article 4: AI Literacy - The Full Text

Providers and deployers of AI systems shall take measures to ensure, to their best extent, a sufficient level of AI literacy of their staff and other persons dealing with the operation and use of AI systems on their behalf, taking into account their technical knowledge, experience, education and training and the context the AI systems are to be used in, and considering the persons or groups of persons on whom the AI systems are to be used.

What This Means in Practice

"Providers and deployers"

This covers two groups: companies that build AI systems (providers) and companies that use them (deployers). If your team uses any AI tool at work, you're a deployer.

"sufficient level of AI literacy"

Staff must understand what AI can do, what it can't do, and how to use it responsibly. The bar isn't expert-level - it's appropriate to their role and context.

"taking into account their technical knowledge, experience, education and training"

Training must be role-appropriate. A developer needs different training than a marketing manager. One-size-fits-all programs don't satisfy this requirement.

"the context the AI systems are to be used in"

Training must cover the specific AI tools your organization actually uses, not generic AI theory. Context matters.

Key Definitions (Article 3)

AI Literacy

Skills, knowledge, and understanding that allow providers, deployers, and affected persons to make an informed deployment of AI systems and to gain awareness of the opportunities and risks of AI and possible harm it can cause.

Deployer

Any natural or legal person, public authority, agency, or other body using an AI system under its authority, except where the AI system is used in the course of a personal non-professional activity.

Provider

A natural or legal person, public authority, agency, or other body that develops an AI system or a general-purpose AI model and places it on the market or puts the AI system into service under its own name or trademark.

Who Needs Training?

Article 4 applies to anyone dealing with the operation and use of AI systems. The training focus should vary by role:

Role	Focus	Key Topics
General Staff	AI literacy fundamentals	What AI is, recognizing AI-generated content, responsible use policies, limitations and risks of AI outputs
AI Operators	Tool-specific competence	Effective prompt engineering, output validation, understanding tool-specific limitations, data privacy in AI interactions
Technical / IT	Risk assessment & implementation	AI risk categories, data handling requirements, system integration, monitoring and evaluating AI outputs, incident response
Management	Governance & strategic oversight	AI governance frameworks, compliance obligations, risk management, organizational AI policies, vendor assessment

Enforcement Timeline (Article 113)

The EU AI Act entered into force on 1 August 2024. Different provisions apply at different dates:

2 February 2025Chapters I & II (including Article 4)

AI literacy requirement, definitions, and scope. Already enforceable.

2 August 2025Prohibited practices, GPAI obligations

Chapter III Section 4, Chapter V, VII, XII, Article 78. Banned AI uses and general-purpose AI model rules.

2 August 2026Full regulation applies

All remaining provisions take effect, including high-risk AI system classification and conformity assessments.

2 August 2027Pre-existing high-risk systems

High-risk AI systems already placed on the market or put into service must comply with all obligations.

Article 4 is already in effect. Waiting for the 2026 deadline means missing the literacy requirement that's enforceable today.

How to Prove Compliance

There is no central EU certification for AI literacy training. Compliance is process-based - you demonstrate it through documentation, not a certificate on the wall.

What Companies Need for an Audit

Training Logs

Records showing who received training, when, and what content they completed. Must cover all staff and persons dealing with AI systems.

Syllabus Mapping

Documentation proving your training content covers the relevant risks, opportunities, and contexts for your organization's specific AI use.

Role-Based Assessment

Documented assessment of staff technical knowledge and experience before assigning training levels - demonstrating you took individual backgrounds into account. This is what the AITutoro diagnostic produces: a dated, role-aware record of who was assessed, on what task, and how they scored - the first document in your compliance file.

ISO/IEC 42001 (AI Management Systems) serves as a gold-standard reference framework for building your compliance documentation.

AI Risk Categories: The Big Picture

The EU AI Act organizes AI systems into four risk tiers. While Article 4's literacy requirement applies broadly, understanding the full framework helps contextualize your training obligations.

Prohibited

AI practices banned outright - social scoring, manipulative systems, real-time biometric identification (with exceptions).

Training: Staff must recognize prohibited uses to avoid deploying them.

High-Risk

AI in critical areas - hiring, credit scoring, education, law enforcement. Subject to conformity assessments.

Training: Operators need deep understanding of risks, monitoring obligations, and human oversight requirements.

Limited Risk

AI systems with transparency obligations - chatbots, deepfakes, emotion recognition must disclose AI involvement.

Training: Staff must understand disclosure requirements and implement them correctly.

Minimal Risk

Most AI applications - spam filters, AI-assisted writing, recommendation systems. No specific obligations beyond Article 4.

Training: General AI literacy as required by Article 4.

The Diagnostic Is Your Article 4 Starting Point

AITutoro is a workforce AI diagnostic - a five-minute, per-employee assessment that scores how well people actually use AI.

•Role-based assessment, documented: every run records each person's role, tools, and verified score on a dated report - exactly the documented assessment of staff knowledge and experience an Article 4 audit file needs, produced before any training is assigned.
•Evidence, not attendance: compliance proof is process-based. A scored diagnostic shows you took individual backgrounds into account; a sign-in sheet does not.
•Then train the gaps: the per-person and per-team gap lists tell you what training each role actually needs. We can deliver that training as an add-on, but the diagnostic record stands on its own.

Frequently Asked Questions

How does the diagnostic relate to the EU AI Act?

Does Article 4 of the EU AI Act apply to my company?

What does 'AI literacy' mean under the EU AI Act?

Is there an official EU certification for AI training?

What documentation proves compliance with Article 4?

When did the AI literacy requirement become effective?

What are the penalties for non-compliance with the EU AI Act?

Can existing corporate training satisfy Article 4?

How does AITutoro help with EU AI Act compliance?

Article 4 Is Already in Effect

Your company's AI literacy obligation is live today. Start building your compliance record now - before someone asks for it.