AI disclosure requirements in the Netherlands: what applies since 2 August 2026
The EU AI Act's transparency obligations are a Union-wide regulation, so the substance is the same in the Netherlands as everywhere in the EU: from 2 August 2026, AI systems that interact with people must disclose they're AI, synthetic media must carry machine-readable marking, and deepfakes and certain AI-generated public-interest text must be labeled. What differs country by country is who enforces it, in what language your users encounter it, and how quickly complaints turn into questions.
Enforcement in the Netherlands
In the Netherlands, supervision of the AI Act is being organised with the Dutch Data Protection Authority (Autoriteit Persoonsgegevens) and the Dutch Authority for Digital Infrastructure (RDI) in coordinating roles, as reported in mid-2026 while the national supervisory structure is finalised. From 2 August 2026, Article 50 falls under the enforcement of the competent national supervisory authorities, with penalty exposure up to €15 million or 3% of worldwide annual turnover under the Act's penalty framework. Practically: the AP already runs an active complaints culture around digital rights in the Netherlands, so assume complaints are a realistic trigger, not a theoretical one.
The four duties, as they show up on your site
- Article 50(1): chatbot disclosure
- Any AI system interacting with visitors must make that clear no later than the first interaction, in a clear and distinguishable way.A line in your terms doesn't satisfy it.
- Article 50(2): machine-readable marking
- AI-generated audio, images, video, and text must be marked in a machine-readable, detectable format.For generative systems already on the market before 2 Aug 2026, this specific duty has a transition until 2 Dec 2026; the rest does not.
- Article 50(3): emotion recognition & biometric categorisation
- People exposed must be informed the system is operating.
- Article 50(4): deepfake and AI-text labels
- Visible disclosure for deepfakes, and for AI-generated or manipulated text published to inform the public on matters of public interest.
the Netherlands-specific practicalities
- Dutch consumers and privacy advocates are unusually complaint-active; the AP's existing GDPR complaint channels are a likely funnel for AI-transparency grievances too
- If your chatbot serves Dutch users in Dutch, make the AI disclosure equally clear in Dutch, a clear notice your users can't read is a weak notice
- The RDI's involvement signals a technical-supervision flavor: machine-readable marking (50(2)) is the kind of duty a telecom/digital-infrastructure regulator knows how to test
- Keep your evidence file bilingual-ready: screenshots of the Dutch-language first-interaction state alongside the English one
Disclosure language should match your user-facing language. For a site serving the Netherlands in Dutch, a first-message line such as "Je chat met een AI-assistent" (you're chatting with an AI assistant) alongside your English notice is the low-cost, high-clarity pattern.
Common questions
We're a US company with Dutch customers, who would even come after us?
Enforcement runs through the national supervisory authorities, and cross-border reach follows the users you serve. The realistic near-term risk isn't a surprise raid, it's a complaint from a Dutch user or competitor that obliges you to show your disclosures were in place. That's an evidence question.
Is there Netherlands-specific guidance we should follow?
The binding text is the EU regulation plus the Commission's guidelines and the Code of Practice; national authorities add procedure and emphasis. Track AP and RDI communications through 2026, and treat the Commission's Article 50 materials as the substance.