In the Sandbox: How AI Governance Frameworks Are Shaping Responsible Innovation

AI and privacy sandboxes are becoming central tools for governments and companies seeking to experiment responsibly with new technologies. The EU is gearing towards the August 2026 deadline for establishing AI regulatory sandboxes. But what does it really mean to create a “safe space” for AI innovation? This session explores how regulatory sandboxes can help bridge the gap between innovation and compliance, enabling real-world testing of AI systems while safeguarding privacy, data protection and fundamental rights. Drawing on experiences from European and national initiatives, our panel will examine how sandboxes operate in practice, which success factors are emerging and how the EU AI Act is shaping this landscape. Participants will gain practical insights on how to join or design a sandbox, manage risks through collaborative governance and turn regulatory flexibility into a driver for trustworthy AI innovation. 

What you will learn: 

• Understand how AI and privacy sandboxes work in practice: what they are, how they are governed and how they support responsible experimentation under the EU AI Act. 
• Identify best practices and common pitfalls when designing or participating in sandbox projects, including data governance, accountability and stakeholder coordination. 
• Discover how to leverage sandbox experience to strengthen company or institutional readiness, enhance collaboration with regulators and scale trustworthy AI.