By Gianluca Misuraca, Special Advisor, Democracy in the Digital Age, Re-Imagine Europa
We are glad to introduce the work of Gianluca Misuraca, Special Advisor on AI and Data Policies for Re-Imagine Europa’s Democracy Task Force. This Expert View, the first in a new series aiming to share expert opinions and insights with the public, is about the future of the governance of AI as well as the future of governance with the help of AI. Those interested in reading a more comprehensive paper, co-authored by Gianluca, can do so here.
Towards leading the digital decade…
With the launch of the Digital Europe Programme and the presentation of the Digital Compass during the conference “Leading the Digital Decade”, held on 1st and 2nd June 2021, the EU is completing the framework for setting the direction to shape Europe’s digital transformation towards 2030.
This process initiated back in April 2018, with the signature by EU Member States of the AI Declaration and the adoption of the Communication on “Building Trust in Human-Centric AI”. The European perspective on AI was further defined in December the same year, with the Communication on “Maximising the benefits for AI in Europe” and the launch of the Coordinated Plan on AI.
Building on the vision outlined by the Commission, during 2019 the High Level Expert Group on AI Ethics developed the Guidelines and a set of Recommendations for Trustworthy AI. These served as an input to structure the “White paper on AI – A European approach to excellence and trust” and the “European Strategy for Data”, presented in February 2020 as the first pillars of the Strategy on Shaping Europe’s digital future set out by the new Commission under the lead of Ursula von der Leyen.
The principles underpinning those Communications were further specified by the President of the Commission during her State of the Union speech in September 2020, and elaborated in the proposed regulation on Data Governance, that aims to foster the availability of data for use by increasing trust in data intermediaries and by strengthening data-sharing mechanisms across the EU. The Digital Services and Digital Markets Acts, further proposed a single set of new rules applicable across the whole EU to create a safer and more open digital space, and offer an alternative European model to data handling practice of major tech platforms.
This set of measures published in November and December 2020 respectively, have been then complemented by the so called “AI Package” presented on 21st April 2021. Three years after the AI Declaration has been signed, the European proposal for a legal framework on AI prepare the ground for proportionate and flexible rules to address the specific risks posed by AI systems and set the highest standard worldwide. At the same time, the 2021 review of the Coordinated Plan on AI outlines the necessary policy changes and investment at Member States level to strengthen Europe’s leading position in the development of human-centric, sustainable, secure, inclusive and trustworthy AI.
As the overarching Communication on “Fostering a European approach to AI” underlines, “Faced with the rapid technological development of AI and a global policy context where more and more countries are investing heavily in AI, the EU must act as one to harness the many opportunities and address challenges of AI in a future-proof manner“. To this end, to promote the development of AI and address the potentially high risks it poses to the safety and fundamental rights equally, the Commission presented both a proposal for a regulatory framework on AI and a revised coordinated plan for Member States to align their strategic efforts and strengthen Europe’s positioning worldwide.
However, while most of the attention of media, experts and policymakers has been focusing on the proposed AI regulation, it is only the combination of the two instruments that can make sure that European citizens can trust what AI has to offer and set the basis for a real European “digital sovereignty”.
The trailblazing role of adopting AI for the public sector…
An important aspect that can be found in the new Coordinated Plan on AI is for sure the prominence given to the public sector. While in the first Plan of 2018 the use of AI to support government operations and stimulate adoption of AI in public services was very limited, and the various national strategies on AI presented a “patchwork” approach with varying scope and depth of policy actions, often reduced to “sermons” on the need to stimulate awareness and share knowledge, with little room for “sticks” and “carrots”, due to lack of proper enforcement rules, funding and incentives, the new Plan ascribes a central and instrumental role to the public sector.
As the section of the policy document emblematically titled “Make the public sector a trailblazer for using AI”, stresses, “For deeper and wider AI uptake to become a reality, Europe’s public sector should have access to adequate funding and be equipped, skilled and empowered to conduct strategic and sustainable purchasing and adoption of AI-based systems“.
In this perspective, it is not only remarked that AI applications can contribute to better public services, for instance by “improving citizen-government interaction, enabling smarter analytical capabilities or improving efficiency across public-sector domains and supporting democratic processes”, as it was already mentioned in the previous Plan and further documented by the growing literature and policy reviews in the field, following the pioneer AI Watch report on AI use and impact in public services in the EU.[1] The new plan instead makes a clear reference to the unprecedented opportunity to accelerate uptake of AI in the public sector that the Recovery and Resilience Facility (RRF) can unleash if appropriate strategic interventions are designed at national level and aligned with the AI strategies.
For this reason, Member States are encouraged to focus on building capacity to seize the advantages of predictive analytics and AI in policymaking and public service delivery and a crucial element for success is identified in relation to public procurement. The innovative use of procurement mechanisms to develop appropriate funding instruments, and if possible impact investment approaches, is key to stimulate adoption of trustworthy and secure AI.
To this end, the Commission announces that will launch an Adopt AI programme to support public procurement of AI systems and transforming procurement processes, also designing a public procurement data space and facilitate Member State peer-learning and knowledge sharing.
In this regard, another innovation is the focus on supporting the adoption of AI in public administrations at local level, which will be funded by the Digital Europe Programme and linked to the Digital Innovation Hubs deployed at regional level across the EU. This can become a fundamental lever for the deployment and scale-up of AI-powered Local Digital Twins and the creation of AI algorithm registries, also encouraging the use of catalogues of AI-enabled applications, for example through the AI-on-demand platform and the Living-in.EU action promoted by the Commission with partner networks of private and public stakeholders.
Back to the future: how to ensure EU leadership in the global governance with, of and by AI?
Within this fast evolving policy context, and potentially changing technological paradigms, the EU has made the choice to address the risks generated by specific uses of AI through a set of complementary, proportionate and flexible rules, with the aim that these rules will also provide Europe with a leading role in setting the global gold standard. Although this framework should intervene only in those cases that existing national and EU legislations do not cover, it aims to give AI developers, “shapers” and users the clarity they need. It is therefore evident that the EU intends to place itself as a player for defining the global governance of AI, promoting the convergence of regulatory approaches at the global level.
In this respect, both the Digital Europe Programme and the Digital Compass presented during the Leading the Digital Decade event, but already adopted in March 2021, represent the cornerstone of the implementation strategy that the EU is setting out to realise the vision for a successful digital transformation of Europe by 2030, which is also critical to achieving the transition towards a climate-neutral, circular and resilient economy.
As declared at the launch of the 2030 Digital Compass “The EU’s ambition is to be digitally sovereign in an open and interconnected world, and to pursue digital policies that empower people and businesses to seize a human-centred, sustainable and more prosperous digital future. This includes addressing vulnerabilities and dependencies as well as accelerating investment.”
Europe’s Digital Compass revolves around four cardinal points and concrete targets, that is to ensure 1) Digitally skilled citizens and highly skilled digital professionals; 2) Secure, performant and sustainable digital infrastructures; 3) The digital transformation of businesses; and 4) The Digitalisation of public services. In this last regard, the Berlin Declaration on Digital Society and Value-based Digital Government, signed on 8 December 2020 by the responsible ministers of all EU Member States, affirms European leaders’ strong commitment to fundamental rights and European values and emphasises the importance of digital public services.
This clear and renewed endorsement of the importance of Digital Government, coupled with the central role that the public sector is to play for the adoption of AI, confirms the need to harnessing the multiple roles of AI for the public sector to guarantee an effective “European digital renaissance”.
This requires not only focusing on the use of AI in government (addressing the governance with AI part of the equation) but also making sure that, especially in the public sector, the regulatory aspects of the governance “of and by” AI are addressed, exploiting the power of digital technologies but safeguarding at the same time citizens for possible misuses and negative consequences of using AI.
Governing through the perils and the power of AI in the public sector is a complex art,[2] which can only be mastered if a shared governance structure is established at the global level, so to ensure that EU Digital Rights and Principles are fully respected in the online space as they are in the real world.
For this reason, the EU is promoting its human-centric digital agenda within International Organisations and through establishing an International Alliance on human-centric AI, aiming at developing strong international digital partnerships, combining EU internal investments with the significant funding available under the new external cooperation instruments so to work with partners around the world in achieving common global objectives.
At the European level, instead, the Digital Europe Programme launched this year is the first-ever financial instrument that dedicates an overall budget of over €7.5 billion to finance projects in five key areas: Supercomputing; AI, cybersecurity, advanced digital skills and the use of technologies across the economy and society. In addition, to accelerate the post-Covid economic recovery, the Commission will facilitate the rapid launch of multi-country projects, combining investments from the EU budget, the Member States and industry, building on the Recovery and Resilience Facility and other EU funding. In their Recovery and Resilience Plans, Member States are in fact committed to dedicating at least 20% to the digital priority and possible multi-country projects include a pan-European interconnected data processing infrastructure; the design and deployment of the next generation of low power trusted processors; and connected public administrations, which can be in fact the heart pump to make sure investment in AI is properly governed and will permit the unprecedented scale and intensity of cooperation that is necessary to achieve a successful digital transformation in Europe.
[1] Misuraca G., and van Noordt, C., 2020. https://publications.jrc.ec.europa.eu/repository/handle/JRC120399
[2] Governing algorithms: Perils and powers of AI in the public sector, May 2021, Lead-Author, Gianluca Misuraca
https://digitalfuturesociety.com/report/governing-algorithms/