Artificial intelligence act and regulatory sandboxes
This briefing by the European Parliamentary Research Service examines the concept of AI regulatory sandboxes as proposed in the European Commission's draft Artificial Intelligence Act. It outlines the expected benefits and concerns surrounding regulatory sandboxes, such as fostering innovation versus risks of regulatory fragmentation or data protection issues. Additionally, the document reviews stakeholder feedback and the European Parliament's positions regarding sandbox implementation, liability protection, and SME access.
Summary and content fields are drawn from the document text and are pending editorial review.
Who issued it
Who it governs
Members of the European Parliament and parliamentary staff seeking background material to assist them in their legislative work on the AI Act.
What the document contains
The briefing covers the definition and role of regulatory sandboxes, their expected benefits and risks, the European Commission's proposals for AI sandboxes under the draft AI Act, and key policy debates involving liability, harmonisation, and data protection.
It does not provide binding legal rules or final text of adopted legislation, as it is an analytical briefing on a legislative proposal.
- Regulatory sandboxes allow businesses to test and experiment with innovative products under a regulator's supervision in a controlled environment.
- The European Commission's draft AI Act proposes coordinated national AI regulatory sandboxes to support innovation and assist small-scale providers and start-ups.
- Key concerns around AI sandboxes include potential misuse, risks to consumer and data protection, and fragmentation of the EU single market.
- Sandbox participants remain fully liable under applicable EU and Member State legislation for any harm inflicted on third parties during testing.
- Stakeholders and experts have raised issues regarding the lack of liability protection, the optional nature of national sandboxes, and alignment with GDPR principles.
· What are regulatory sandboxes? · Expected benefits and concerns · European Commission AI act proposal · Key policy issues for discussion
How it scores
Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.
Fairness
2/5Non-discrimination and equitable treatment
Experts have alerted to the potential negative impacts on consumer protection brought about by this 'race to the bottom' in the fintech sector.
Expected benefits and concerns
screening score · 2026-10-23
Human Dignity
0/5Respect for the inherent worth of persons
screening score · 2026-10-23
Privacy
3/5Protection of personal data and private life
Moreover, further processing of personal data (i.e. for a purpose other than that for which such data have been collected) would be allowed in the AI regulatory sandbox
Data protection
screening score · 2026-10-23
Transparency
3/5Openness about how AI systems work and are used
Its April 2022 committee draft report on the AI act argued for more transparency on the implementation and use of AI sandboxes.
SUMMARY
screening score · 2026-10-23
Impact Assessment
3/5Required assessment of an AI system's risks/impacts before or during use
Their objective would be to foster AI innovation by establishing a controlled experimentation and testing environment for innovative AI technologies, products and services during the development phase
Purpose
screening score · 2026-10-23