AI Act
The EU AI Act (Regulation (EU) 2024/1689) is the European regulation that lays down rules for placing AI systems on the market and using them, classified by risk: prohibited practices, high-risk systems, transparency obligations and rules for general-purpose AI models. It entered into force on 1 August 2024 and has applied in general since 2 August 2026. Regulation (EU) 2026/1744 moved the high-risk obligations to 2 December 2027 and 2 August 2028.
Source: Regulation (EU) 2024/1689 (AI Act) Articles 4, 5, 27, 99, 111 and 113 and Annex III; Regulation (EU) 2026/1744 (Digital Omnibus on AI); Dutch parliamentary paper 22 112, no. 4318 (2026)
The Act applies in phases. Prohibited practices and AI literacy have applied since 2 February 2025, the rules for general-purpose AI models since 2 August 2025 and most other provisions since 2 August 2026. The Digital Omnibus on AI (Regulation (EU) 2026/1744, in force since 27 July 2026) postponed the high-risk obligations: to 2 December 2027 for the use cases in Annex III and to 2 August 2028 for AI in products under Annex I. From 2 December 2026 a ninth prohibition applies, to AI systems that generate or manipulate non-consensual intimate imagery of identifiable persons or child sexual abuse material. Since the omnibus, Article 4 requires measures that support the AI literacy of staff, without any specific level having to be ensured.
In the financial sector, the main high-risk systems are those used to evaluate the creditworthiness or establish the credit score of natural persons, except for fraud detection, and those for risk assessment and pricing in life and health insurance (Annex III, point 5(b) and (c)). Deployers of such systems carry out a fundamental rights impact assessment before first use (Article 27). The Act applies directly, but national supervision follows from the Dutch AI Act Implementation Act (Uitvoeringswet AI-verordening), which the government, in August 2026, planned to submit in the second quarter of 2027; under the proposal, the Dutch Authority for the Financial Markets (AFM) and the Dutch Central Bank (DNB) supervise AI at financial institutions.
Fines can reach EUR 35 million or 7 per cent of worldwide annual turnover for prohibited practices and EUR 15 million or 3 per cent for most other obligations (Article 99 of Regulation (EU) 2024/1689). For the board, compliance starts with an AI inventory that records each system's role, provider or deployer, and risk category. The Act sits alongside the GDPR: anyone processing personal data with AI applies both, and the fundamental rights impact assessment may refer to the data protection impact assessment (Article 27(4)).
Last updated: · Editor: ONE Risk Advisory
Frequently asked questions
When do the obligations for high-risk AI apply?
Following the Digital Omnibus on AI, they apply from 2 December 2027 to the use cases in Annex III, such as credit scores of natural persons and pricing in life and health insurance, and from 2 August 2028 to AI in products under Annex I. A system already on the market before then is only covered after a significant change in its design; systems used by public authorities must comply by 2 August 2030.
Who supervises the AI Act in the Netherlands?
That will be laid down in the Dutch AI Act Implementation Act, which was still being prepared in August 2026. The government proposal of April 2026 would designate ten market surveillance authorities, including the Dutch Data Protection Authority (Autoriteit Persoonsgegevens, AP), the RDI, the AFM and DNB; the RDI becomes the central contact point and coordinates together with the AP. AI used by financial institutions for financial services falls to the AFM and DNB. The obligations apply even without the implementation act, because the regulation applies directly.
What is the difference between the AI Act and the GDPR?
The GDPR protects individuals when their personal data is processed, whatever the technology. The AI Act sets requirements for AI systems and their use according to the risk of the application, even where no personal data is involved. For a high-risk system that processes personal data, both apply; the fundamental rights impact assessment may then refer to the data protection impact assessment (Article 27(4)).