Is a free AI course enough for Article 4? Sometimes, but usually not

Article 4 of the EU AI Act has applied since 2 February 2025 and requires organisations to ensure a sufficient level of AI literacy among people working with AI systems on their behalf. The benchmark is not that everyone completed a course, but that measures fit technical knowledge, experience, education and the context in which the AI systems are used. There is no separate fine attached to Article 4. The Digital Omnibus has been formally adopted but not yet published; the adopted text changes Article 4 into a duty to take measures that support the development of AI literacy. The current text remains in force until the amendment enters into force. Afterwards, clients, auditors, or supervisory authorities can still ask for substantiation of the measures taken.

Free AI courses are a good starting point for individual curiosity and general basics about prompts, capabilities and limitations. A free course can form part of appropriate measures, but suitability depends on knowledge, experience, context, people affected, and actual AI use, not organisation size. The friction sits with roles that already use AI operationally: recruiters screening candidates, healthcare professionals working with decision support, staff in financial services, teachers, and policy makers in government. For those roles Article 4 asks for content that matches their context and risks, and it is exactly that link between role, risk and training that generic free courses lack.

Evidence means an organisation can show who completed which training, how understanding was assessed, and which role it matched. An open course does not provide that context automatically. LearnWize records it in an organisation-wide dossier with learner records. Selected evidence changes and exports receive a traceable audit chain; content is updated periodically under agreed arrangements.

If you opt for a programme, there are three delivery models. Co-brand: the programme runs under your name alongside the platform, useful for training providers and partners. For white-label delivery, brand and delivery scope are agreed in advance; removing every LearnWize reference, including certificate and registry branding, is not a standard promise. SCORM 1.2: you run the modules in your own LMS; LearnWize delivers packages tested end-to-end in Moodle, while other LMSs depend on their own SCORM 1.2 implementation. When to choose what: stay with free courses if your AI use is minimal and you can substantiate that choice, choose SCORM if your LMS and reporting are already in order, and choose co-brand or white-label if you want testing, certificates and an evidence file without building anything yourself.