AI Act Impact on the Public Sector: Implementing and Enforcing the AI Act

For the public sector, preparing for the implementation of this groundbreaking legislation is essential, yet it comes with challenges, such as the need to recruit digitally-savvy professionals and gain a deeper understanding of AI. This part of our blog post series explores the unique characteristics of the AI Act and examines the measures, which the public sector can take to ensure its readiness for its effective implementation, while addressing the hurdles it is currently facing.

The AI Act represents a novel approach to governing and overseeing artificial intelligence, introducing a fresh perspective for both AI innovators and regulators

As the AI Act's subject, the public sector bears the responsibility of serving as its implementer and enforcer. Additional investments will be required for AI governance and oversight for setting up the needed skills, processes and infrastructures as a new field of expertise for the public sector.

One of the pivotal requirements for the public sector outlined in the AI Act is the establishment of regulatory sandboxes. Regulatory sandboxes are controlled environments, which are envisioned to be the cornerstone of implementing and refining the legislation. These offer a unique space where the public sector can collaborate closely with AI innovators and various stakeholders. The primary aim is to facilitate the practical application of the AI Act's requirements, while supporting innovation. Experiences gained in these regulatory sandboxes will play a vital role in refining and adapting the AI Act implementation to real-world scenarios. This innovative approach fosters a dynamic learning environment, enabling policymakers, AI developers, and regulators to work collaboratively and iteratively. As they navigate the uncharted waters of AI governance, the insights and lessons learned from these sandboxes will contribute to the ongoing improvement of the AI Act.

The public sector is also responsible for the enforcement of the AI Act. Regulators must ensure that AI systems and actors across the AI value chain adhere to the requirements of the AI Act. This will require exercising oversight, control, providing appropriate guidance and accrediting auditing and certification bodies to assess and verify compliance with future EU harmonised standards. 

One of the most urgent challenges for the public sector as a regulator is ensuring sufficient investments in the upskilling of its existing staff or attracting new experts who are able to effectively supervise AI systems. This includes understanding the nuances of AI technology, staying abreast of evolving regulations, and developing the capacity to evaluate AI system compliance. Therefore, investing in staff training and development becomes a priority to meet these formidable challenges effectively.

As this significant law is put into action, it represents a big step in dealing with the new challenges surrounding AI. It will contribute to AI being developed and used in a responsible and ethical way, while aiming to protect the rights, safety, and the overall well-being of people and society.

Your next steps

The following steps will help the public sector to successfully navigate the challenges posed by the AI Act and for the public sector to fulfil its role as a regulator of AI:

1. Begin by ensuring the best placed public sector organisations to carry responsibility for the AI Act. These may be organisations with expertise in data protection, digitalisation, cybersecurity and others.
2. Map sectoral regulators who would need to be involved in the implementation of the AI Act in specific sectors, e.g. banking, energy and water utilities, transportation. Plan how to bring together different public sector actors and encourage them to coordinate.
3. Create a plan for tackling any missing capabilities, tools or processes to successfully run a regulatory sandbox programme and enforce the AI Act requirements. Build relationships with industry stakeholders and innovators.
4. Invest in training and education programs to ensure staff is well-versed in AI technology, the AI Act's requirements, and the evolving regulatory landscape. Consider the AI literacy requirements in the AI Act. 

This article is part of a series called “AI Act impact on the public sector”. For further exploration of the AI Act's risk-based approach in the public sector, please refer to the subsequent article here. Additionally, further exploration of the public sector responsibilities as an AI provider and deployer, see the following part of the series.This series does not attempt to provide any legal analysis or interpretation.