EU sets rules for AI regulatory sandboxes, with free access for start-ups
The EU has adopted Commission Implementing Regulation (EU) 2026/2487 of 10 November, which sets the rules for AI regulatory sandboxes under the AI Act, Portuguese law firm Morais Leitão reports. The European Commission regulation covers how sandboxes are set up, run and supervised, including access, confidentiality, data protection, risk management and recognition of test results. Member states must have national sandboxes in place by 2 August 2027.
Small and medium-sized companies, including start-ups, get priority and free access to regulatory guidance, advice and supervision, though exceptional costs can still be charged if they are announced in advance and recovered fairly. Providers can apply to sandboxes in other member states, not only where they are based, and applicants from outside the EU must pay a fee and appoint an authorised representative in the Union. Systems not yet on the market are eligible, as are existing systems undergoing substantial changes.
Selected participants agree a sandbox plan with the authorities covering timetable, objectives, legal requirements to be assessed and handling of personal data. Within 45 working days after a project ends, they receive an exit report that market surveillance authorities and notified bodies across the EU must recognise, though it does not count as a declaration of conformity.
Testing can be suspended if risks to health, safety or fundamental rights cannot be mitigated, and taking part does not relieve companies of their duty to comply with the law, according to Morais Leitão.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like AI Policy Brief, AI in Europe or Daily AI Brief.
Also, consider following us on social media:
More from Regulation
Nov 18 San José pauses data centre approvals while it writes new standards Nov 18 Baltimore teens sue schools, police and Omnilert after AI mistook a chips bag for a gun Nov 18 California report urges state power to shut down dangerous AI models and cap self-improving AI Nov 18 Anthropic, OpenAI and Mistral sign Paris Call to be able to stop their own AI models Nov 18 Mistral is not Macron's "personal project", CEO Arthur Mensch saysAI in Europe
Daily AI news from Europe, chosen by Rene van Pelt: European companies, the EU and the UK, chips and data centres, deals and regulation, in a few minutes.
Industry analysis
2025 Global Business Services Agenda: Gen AI Takes Center Stage
This industry analysis by The Hackett Group explores the transformative impact of generative artificial intelligence (Gen AI) on global business services (GBS) in 2025. The study highlights the shift from exploration to acceleration of Gen AI initiatives, with 89% of executives advancing these projects to improve customer satisfaction, innovate products, and reduce costs. The report also discusses the challenges and strategies for successful Gen AI adoption, emphasizing the need for a technology-enabled operating model and the importance of reskilling the workforce.
Read moreYou may also like
EU seeks feedback on draft AI Act fundamental rights impact assessment template
EU Publishes First Harmonised Technical Standard Under AI Act
Italy's AI Act Decree Takes Effect October 23
EU science and ethics advisers say AI needs rigorous safeguards and rules that keep pace
EU Digital Fairness Act Proposal Gives Consumers Rights When AI Agents Shop for Them
Daily AI Brief: the AI news that matters, in your inbox.