A practical guide to understanding the EU Artificial Intelligence Act, reading the relevant duties and turning them into a work plan for your organisation.
Updated 26 October 2026. The AI Act is Regulation (EU) 2024/1689, which sets harmonised rules for artificial intelligence in the European Union. It does not ban every use of AI. Its approach is risk-based: the more a system may affect health, safety or fundamental rights, the stronger the requirements become.
How to read the Act
Start with the definitions and your organisation’s role. The Regulation distinguishes, among others, providers that develop or place a system on the market and deployers that use it under their authority at work. A business can have different roles depending on the system and whether it modifies or places it on the market.
Then map the use case against the Articles and Annexes: check prohibited practices, possible high-risk classification, transparency duties towards people and specific requirements for general-purpose AI models. The question is not only “which tool do we use?” but what it does, in which workflow, with what data and with what impact.
Dates: check for updates
As of 26 October 2026, prohibitions and the AI literacy requirement have applied since 2 February 2025; obligations for general-purpose AI models have applied since 2 August 2025. Most provisions, including Article 50 transparency rules, apply from 2 August 2026. The amending Regulation (AI Omnibus) entered into force on 27 July 2026 and changed the timetable for certain categories: the Commission indicates 2 December 2027 for several high-risk-system rules and 2 August 2028 for systems embedded in regulated products. Check the consolidated text and official guidance for your case: dates depend on the category and may be clarified through further measures.
A practical route for SMEs and teams
- Make an inventory. List AI tools, suppliers, users, tasks and workflows, including AI features embedded in software you already buy.
- Describe use cases. For each one, note purpose, affected people, data, human oversight, outputs and decisions that rely on them.
- Clarify roles and risk. Compare the use case with definitions, prohibitions, transparency duties and Annexes. Ask suppliers for documentation and instructions.
- Set usage rules. Define approved tools, human checks, data limits, error handling and ways to report incidents or questions.
- Train people and keep evidence. AI literacy should fit people’s skills, context and use of systems; keep proportionate records of training, decisions and reviews.
How I can help as a consultant
I can help map tools and workflows, turn business needs into a practical checklist, coordinate marketing, operations and IT, prepare operating procedures and training materials, and support an implementation plan with clear owners and dates. For questions requiring legal interpretation, impact assessments or specialist duties, I involve or refer to qualified legal and technical professionals. Organisational consulting does not replace legal advice.
Where to start this week
Pick one concrete workflow, such as content generation or lead triage. Write down which system is used, who checks the output and what happens if it is wrong. This first-use-case sheet makes it easier to ask suppliers the right questions and decide which specialists to involve.
Official sources
- Regulation (EU) 2024/1689 — EUR-Lex text
- European Commission — European AI regulatory framework
- European Commission — AI Act FAQ and application timeline
This article provides informational and organisational guidance, not legal advice. Check the current text, implementing measures and guidance from relevant authorities.