Regulatory Sandbox

A time‑bounded test environment operated under supervisory arrangements that allows controlled piloting of AI applications with specified oversight, risk controls and temporary regulatory flexibilities to evaluate safety, impacts, and liability allocation prior to full deployment.

A formally managed program or environment administered under SDAIA where companies and agencies can pilot AI systems or privacy-enhancing technologies under relaxed or tailored regulatory conditions, with oversight, reporting requirements, and time-limited permissions to evaluate safety, compliance, and commercial viability prior to scaled deployment.

A national experimental framework allowing providers/operators to develop and trial innovative AI systems under regulatory supervision with tailored oversight, reporting obligations, eligibility criteria and exit conditions to facilitate safe innovation while ensuring compliance and learning for regulators.

A supervised testing framework established under the Act and managed by the Ministry for Digital Transformation that permits providers or deployers to trial AI systems under defined conditions, oversight, reporting and safeguards to support innovation while ensuring compliance with safety and legal obligations.

A restricted, supervised environment where AI developers can test systems—particularly those with medium or high human rights impact—against anticipated regulatory requirements, receive guidance and feedback, and refine compliance measures before full market deployment, without immediate legal penalties for non-compliance.

A supervised testing environment where AI developers may evaluate systems (restricted to those with medium or high human rights impact) against anticipated regulatory requirements, receive structured feedback from the Expert Committee, and iterate designs prior to full market deployment; designed as a learning rather than enforcement mechanism during Stage One.

A supervised pilot environment that allows developers and public-sector actors to test AI solutions under controlled conditions with limited regulatory exemptions and oversight, enabling experimentation, evaluation of impacts, and iterative improvement prior to broader deployment.

A sector‑specific, state‑mandated experimental framework where providers and operators can test AI systems under supervised conditions with tailored regulatory support and oversight, enabling innovation while ensuring compliance with safety, transparency and risk‑management obligations laid out in the implementing measures.

A regulatory sandbox is a supervised, time‑limited environment established by authorities (including privacy‑focused sandboxes) in which innovators can pilot novel AI applications under tailored oversight and temporary legal/regulatory clarification to assess risks, compliance needs, and real‑world impacts before broader deployment.

A controlled environment where organisations can test innovative products or services with regulatory oversight and tailored guidance; used to pilot privacy-friendly AI and digitalisation projects while clarifying how existing data protection law applies and how to operationalise compliance.

A formally authorised testing environment established by national authorities where providers can experiment with AI systems under regulatory supervision and tailored conditions to enable innovation while ensuring oversight, reporting and safeguards for safety and fundamental rights.

A framework set up by a regulator that allows small-scale, live testing of new innovative products, services, business models or delivery mechanisms in a controlled real-world environment. The Hanoi plan encourages such sandboxes for new technologies and services, with an emphasis on transparency, risk management, and public interest protection.

A framework established by a regulator to allow the testing of novel technologies and business models in a controlled environment, often with temporary waivers or relaxations of regulatory requirements, to foster innovation.

A controlled environment created by an experimental legal regime where new technologies and business models can be tested under relaxed regulatory requirements, with appropriate safeguards, to gather data and inform future permanent legislation. This approach allows innovation to flourish without compromising public safety or existing legal principles.

A framework that allows companies to test innovative AI applications under regulatory supervision with certain legal relaxations.

A framework allowing companies to test AI systems under reduced regulatory oversight to foster innovation while maintaining basic transparency.

A framework that allows providers to develop and test innovative AI systems under regulatory supervision before they are placed on the market.

A regulatory sandbox is a framework set up by a regulator to allow small-scale, live testing of new products, services, business models, or delivery mechanisms in a controlled environment. It enables innovators to experiment without immediate full regulatory burden, while regulators gather insights to inform future policy.

Regulatory sandboxes, or AI real-world laboratories, are environments promoted by the KI-MIG where innovative AI systems can be tested under controlled, real-world conditions. They aim to reduce regulatory burdens for innovators while maintaining safety and ethical standards, fostering trustworthy AI development.