Dutch Regulator Sets a 22 July Signing Deadline as EU AI Transparency Rules Bite 2 August
Regulatory summary: The Autoriteit Persoonsgegevens, the Netherlands' coordinating algorithm regulator, is urging providers and deployers of generative AI to sign the EU transparency Code of Practice before the first-signatory list closes on 22 July 2026, ahead of the four AI Act Article 50 duties applying from 2 August.
By Anthony Guerriero, Founder, The Leveraged Years · Reviewed by The Leveraged Years Editorial Desk · Published July 9, 2026 · Last updated July 9, 2026
The Leveraged Years AI Regulation Tracker
Key takeaways
The AP moved from general awareness messaging to a dated, operational instruction: with the Commission's Code now published, providers and deployers should study it and sign before the first-signatory window closes on 22 July 2026, so they are ready when Article 50 applies on 2 August 2026. It also set out, in one place, the four transparency duties and the specific Article 50 sub-paragraphs behind each.
Product and compliance leads at companies shipping chatbots and generative AI to EU users; marketing and media teams producing AI-generated or AI-edited content; deployers running emotion-recognition or biometric-categorisation tools; and legal and privacy officers mapping AI Act readiness for Dutch and pan-EU operations.
Status: AP advice published 9 July 2026.
Inventory your generative and interactive AI systems against the four Article 50 duties this week, decide whether to sign the whole Code or only the relevant section, and complete signing before 22 July 2026 at 18:00 if first-signatory status matters, with disclosures live by 2 August 2026.
Date
Jurisdiction
Rule
Affected professionals
Status or effective date
2026-07-09
Netherlands
The AP moved from general awareness messaging to a dated, operational instruction: with the Commission's Code now published, providers and deployers should study it and sign before the first-signatory window closes on 22 July 2026, so they are ready when Article 50 applies on 2 August 2026. It also set out, in one place, the four transparency duties and the specific Article 50 sub-paragraphs behind each.
Product and compliance leads at companies shipping chatbots and generative AI to EU users; marketing and media teams producing AI-generated or AI-edited content; deployers running emotion-recognition or biometric-categorisation tools; and legal and privacy officers mapping AI Act readiness for Dutch and pan-EU operations.
AP advice published 9 July 2026. Commission Code of Practice published 10 June 2026 and open for signature; first-signatory list closes 22 July 2026 at 18:00. Article 50 duties apply from 2 August 2026.
Frequently Asked Questions
Is the EU transparency Code of Practice legally binding?
No. The AP describes the Code of Practice on Transparency of AI-Generated Content, published by the European Commission on 10 June 2026, as voluntary. The binding obligations come from Article 50 of the EU AI Act, which applies from 2 August 2026.
What happens on 22 July 2026?
22 July 2026 at 18:00 is the cut-off to appear on the published list of first signatories to the Code. It is not the date the law takes effect. Organisations can still sign the Code after that date.
What actually becomes obligatory on 2 August 2026?
From 2 August 2026, the four Article 50 transparency duties apply: AI-interaction disclosure (Article 50(1)), emotion-recognition and biometric-categorisation notice (Article 50(3)), machine-readable marking of AI-generated content (Article 50(2)), and visible or audible deepfake labelling (Article 50(4)).
Does signing the Code replace the legal duties?
No. Signing shows how an organisation intends to meet the transparency requirements. It does not create or discharge the Article 50 obligations, which apply whether or not an organisation signs.
Who is the AP in this context?
The Autoriteit Persoonsgegevens is the Dutch data protection authority and the country's coordinating supervisor for algorithms and AI. Its Algorithm Coordination Directorate is preparing for the AP's oversight role under the AI Act and issued this advice on 9 July 2026.
