The EU Just Standardised What an "AI Professional" Is. Here's What It Means for Your Next Hire

Key Takeaways

• CEN-CENELEC published CWA 18398 in June 2026, defining 37 AI professional roles across seven families and mapping each one to established European competence and qualification frameworks.

• The agreement gives Europe a shared language for hiring, training, and governing AI staff, replacing today's inconsistent and interchangeable job titles.

• It is directly connected to the EU AI Act: the AI Compliance Officer and AI Educator roles map to legal obligations, including the Article 4 duty to ensure staff AI literacy.

• For most businesses, the practical outcome is a small data department — a handful of people covering data, delivery, and governance roles, not 37 separate hires.

• A role is not the same as a job title: most organisations already run AI roles embedded inside existing jobs, whether or not the word "AI" appears on the org chart.

Introduction

Europe now has an agreed answer to a question most businesses have been guessing at: what an AI professional actually is, and which roles a company needs to build and run AI responsibly. That answer arrived in June 2026, and it will shape how you hire, train, and structure teams for the next decade. Here is what changed and what to do about it.

What Is CWA 18398?

CWA 18398:2026 is a CEN Workshop Agreement, published by CEN-CENELEC, the European Committee for Standardization. It was developed in part through ARISA, the Artificial Intelligence Skills Alliance, funded under the European Commission's Erasmus+ programme. A CEN Workshop Agreement is a consensus reference document rather than a binding standard, agreed by interested parties across the 34-plus CEN member countries, including Malta.

The document defines 37 AI professional role profiles, organised into seven families: Data Processing and Analysis, Development, Operations, Support, Guidance, Management, and Governance. Each family maps to a phase of the AI system lifecycle, from preparing data through to overseeing systems in production.

Every role comes with a defined mission, tasks, deliverables, and required competences, drawn from the European e-Competence Framework (e-CF), the EU-agreed scheme already used to describe ICT jobs. Each role is also paired with an educational profile that expresses the learning outcomes needed, levelled against the European Qualifications Framework (EQF). In plain terms, the document links what a role does, what skills it needs, and how to train for it, in one connected structure.

Why It Matters

The value here is not a new idea about AI. It is a common vocabulary arriving at the moment the market most needs one. Europe faces a widening digital skills gap, projected to grow from around 8 million to 20 million ICT professionals needed by 2030. At the same time, the EU's 2030 targets call for more than 90 percent of small and medium-sized enterprises to reach a basic level of digital intensity, and 75 percent of companies to use cloud computing, big data, or AI. Businesses cannot plan a workforce they cannot describe consistently.

CWA 18398 fixes that description problem. When a Maltese accounting firm, a hospitality group, and an iGaming operator all use the same role definitions, hiring, training, and outsourcing become comparable across the market. Job adverts, university courses, and certifications can align to the same reference, which reduces the guesswork in every AI hire.

It also elevates governance to a first-class career track. The framework treats compliance, risk, audit, and responsible-AI roles as peers of engineering roles, not afterthoughts. This reflects regulatory reality. The EU AI Act's Article 4 already requires providers and deployers of AI systems to ensure a sufficient level of AI literacy among staff, and the document names the AI Compliance Officer and AI Educator as the roles most directly tied to those duties.

The most useful concept for a business leader is that a role is not a job title. A person formally titled "information security specialist" can hold the AI Security Specialist role. A financial controller can hold the AI Business Analyst role. The absence of "AI" from a job title does not mean the absence of AI responsibilities. Most organisations already have AI roles distributed across existing staff and simply have not named them.

The Small Data Department Every Business Will Soon Need

CWA 18398 makes it clear that small organisations do not need 37 people. They need a small cluster of roles, often held by two to five individuals wearing several role hats, supported by fractional or outsourced expertise for the parts they cannot justify hiring full-time.

A practical near-future data department for a mid-sized business looks like this:

• Data foundation. A Data Engineer and a Data Analyst or Data Scientist make company data usable and turn it into decisions. This is the core that every other role depends on.

• Delivery. An AI Engineer or AI Application Developer builds and integrates the tools, adding MLOps capability only once systems run at scale.

• Governance and literacy. Someone holds the AI Compliance Officer, AI Governance Officer, and AI Educator roles, ensuring systems are lawful and staff are AI literate under the AI Act. In smaller firms this is frequently fractional or advisory.

• Leadership. A Chief AI Officer or AI Transformation Lead aligns the work to business strategy. For an SME, this is typically a fractional or consultancy role rather than a full-time executive.

The function of this department is straightforward: turn business data into decisions and automation, safely and in compliance. In a large enterprise these roles are separate people. In a small business they collapse into a few individuals, with governance and leadership often supplied by external advisors. Either way, the shape is now defined.

What Businesses Should Do Next

1. Map your existing staff to the 37 roles. Identify the AI roles already embedded in current jobs before assuming you need new hires. Most businesses find they hold several roles unnamed.

2. Define your minimum viable data department. Decide which two to five roles you need first, using the data foundation, delivery, governance, and leadership cluster above as a starting template.

3. Assign the AI Act roles now. The Article 4 AI literacy obligation is already in force. Name who holds the AI Compliance Officer and AI Educator roles, even on a part-time or fractional basis.

4. Use the educational profiles to plan training. Build or buy training against the framework's EQF-levelled learning outcomes so skills are comparable and portable.

5. Decide build versus fractional for governance and leadership. These are the roles SMEs most need and least can hire full-time, and the clearest candidates for external advisory support.

Frequently Asked Questions

What is CWA 18398:2026?

CWA 18398:2026 is a CEN Workshop Agreement published by CEN-CENELEC in June 2026 that defines 37 AI professional role profiles across seven families, each with required competences and a paired educational profile. It is a consensus reference document, not a binding standard, but it is linked to the EU e-Competence Framework, the European Qualifications Framework, and the EU AI Act, which gives it broad practical influence.

Does CWA 18398 mean my business needs to hire 37 different people?

No. The framework is a reference taxonomy, not a hiring quota. Small and medium businesses typically need a cluster of two to five roles, often held by the same individuals across data, delivery, and governance functions, with leadership and compliance roles frequently supplied on a fractional or outsourced basis.

How does CWA 18398 relate to the EU AI Act?

The AI Act assigns legal obligations to organisations, while CWA 18398 defines the professional roles that carry out those obligations. Two roles map most directly: the AI Compliance Officer, accountable for AI Act compliance, and the AI Educator, responsible for the staff AI literacy required under Article 4, which is already in force.

Is a formal "AI" job title required to hold an AI role?

No. The document is explicit that a role is defined by the responsibilities a person fulfils, not by their job title. An employee whose formal title makes no mention of AI can still hold a defined AI role if their tasks involve planning, building, running, enabling, managing, or governing AI systems.

Why does a non-binding agreement matter for businesses?

Its influence comes from its connections rather than its legal status. Because CWA 18398 maps to the e-Competence Framework, the European Qualifications Framework, ESCO, and the EU AI Act, it is positioned to become the shared reference that employers, educators, and certification bodies use to describe AI work across Europe.

Conclusion

The guesswork phase of AI hiring is ending. Map the roles you already hold, define the small data department you actually need, and name your AI Act roles before the literacy obligation catches you unprepared.

Transparency Disclosure: AI-Assisted Content

This article, including any images, was generated with the assistance of a Large Language Model (LLM) but has undergone a comprehensive process of human review and editorial control. In accordance with the exceptions outlined in Article 50(4) of the EU AI Act and the draft Code of Practice, this publication is subject to the editorial responsibility of Synerf. The review process involved verifying factual accuracy, ensuring contextual relevance, and exercising organizational oversight to maintain the integrity of the information provided.

References

[1] AI Professional Role Profiles and Educational Profiles, CEN-CENELEC Workshop Agreement CWA 18398:2026, European Committee for Standardization, Brussels, Belgium, Jun. 2026.

[2] Artificial Intelligence Skills Alliance (ARISA), Erasmus+ Programme — Alliances for Sectoral Cooperation on Skills, European Commission, 2022–2026. [Online]. Available: https://www.aiskills.eu

[3] e-Competence Framework (e-CF) — A Common European Framework for ICT Professionals in All Sectors — Part 1: Framework, EN 16234-1:2019, European Committee for Standardization, Brussels, Belgium, 2019.

[4] European Commission, The European Qualifications Framework: Supporting Learning, Work and Cross-Border Mobility. Luxembourg: Publications Office of the European Union, 2018.

[5] European ICT Professionals Role Profiles — Part 1: 30 ICT Profiles, CEN Workshop Agreement CWA 16458-1:2018, European Committee for Standardization, Brussels, Belgium, 2018.

[6] Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act), Official Journal of the European Union, L series, 12 Jul. 2024. [Online]. Available: http://data.europa.eu/eli/reg/2024/1689/oj

[7] European Commission, European Skills, Competences, Qualifications and Occupations (ESCO). [Online]. Available: https://esco.ec.europa.eu/en

[8] OECD, "OECD Framework for the Classification of AI Systems," OECD Digital Economy Papers, no. 323. Paris, France: OECD Publishing, 2022. [Online]. Available: https://doi.org/10.1787/cb6d9eca-en

[9] Decision (EU) 2022/2481 of the European Parliament and of the Council of 14 December 2022 establishing the Digital Decade Policy Programme 2030, Official Journal of the European Union, L 323, pp. 4–26, 19 Dec. 2022.