On February 19th, the European Commission’s Executive Vice President for A Europe fit for the digital age, Margrethe Vestager, and Commissioner for the Internal Market Thierry Breton unveiled the European Commission’s long-term digital agenda together with two actions plans on data and artificial intelligence (AI).
When read altogether, it appears that the Commission does not wish to precipitate and overhaul rules concerning AI or online platforms. It has however identified the need to level the playing field in digital markets and has presented a wide array of proposals to do that.
The package of the Commission includes the following papers:
- A Communication on “Shaping Europe’s digital future”, which recalls the legislative and non-legislative initiatives that will support the Commission’s strategy in the digital area (hereafter the “EU digital strategy”);
- A Communication on “A European strategy for data”, which presents the Commission’s plans to boost data access, use and re-use and to create a level-playing field in the data space (hereafter the “EU data strategy”);
- A White Paper on Artificial Intelligence, which presents the Commission’s plans to support investments in AI and its ongoing assessment of the legislation addressing AI risks (hereafter “AI white paper”).
- Their release was accompanied by a Report on business-to-government data sharing for the public interest and a Report on the safety and liability implications of AI, the Internet of Things and robotics.
Both the EU data strategy and the AI white paper will be open for public consultation until May 19th, 2020.
Please find below the key takeaways of the Commission’s strategy and action plans.
1. Key takeaways of the EU digital strategy
The digital strategy is underpinned by the three main priorities: raising trust in new technologies, boosting competition in digital markets and preserving a democratic and sustainable society. These priorities will be declined through policies in the following areas of interest:
- Competition: The Commission will explore the need for ex-ante rules in competition that “ensure that markets characterized by large platforms with significant network effects acting as gate-keepers” remain competitive. It will also launch a sector inquiry on new and emerging markets which have not been decided upon yet, according to Commissioner Vestager;
- Platforms’ regulation: As part of the upcoming Digital Services Package, the Commission will propose measures aimed at “harmonizing the responsibilities of online platforms and information service providers and reinforce the oversight over platforms’ content policies in the EU”. It should be noted that the Commission does not mention a review of liability rules within the e-Commerce directive;
- Consumer protection: A new Consumer Agenda focused on empowering consumers to make informed choices will be released end 2020;
- Cybersecurity: The Commission will present a new European cybersecurity strategy that will feature the establishment of a “joint Cybersecurity Unit” and the review of the Security of Network and Information Systems (NIS) Directive;
- Skills: The Commission has committed to reinforce the Skills Agenda as well as the Youth Guarantee to put a strong focus on digital skills in early career transitions (Q2 2020);
- Industrial strategy: On March 10th, the Commission should unveil an Industrial Strategy Package putting forward a range of actions for greening EU industries and levelling the playing field for SMEs.
- Taxation: The Commission will build on the work done at the OECD level and present its plans in a future communication.
- Financial sector: The Commission will take measures for regulating crypto assets and improve operational and cyber resilience in the financial sector. In Q3 2020, it will release a strategy to support pan-European digital payment services and solutions.
- Media and audiovisual sector: In Q4 2020, the Commission should put forward an Action Plan to stimulate access to quality content and media pluralism.
- Environment and climate change: Initiatives will be taken to make data centres climate neutral by 2030. A circular electronics initiative should also promote the reuse and recycling of electronic components of technological devices.
- Earth observation: As of 2021, the Commission will implement a “Destination Earth, initiative” to develop a high precision digital model of Earth. The latter would seek to improve Europe’s environmental prediction and crisis management capabilities.
2. Key takeaways of the data strategy
The data strategy ultimately seeks to create a single market for data. In order to achieve this goal, the Commission plans to take the following steps:
• Presenting a Data Act regarding the voluntary data sharing among businesses, in 2021. It should address issues related to usage rights for co-generated data and allocation of responsibilities. The intellectual property rights framework could also be reviewed;
• Revising competition rules and practices: The Commission should provide guidance to stakeholders on the compliance of data sharing and pooling arrangements through an update of the Horizontal Co-operation Guidelines and individual assessments. It will also adapt its merger control and consider the need to move to an ex-ante regulation for big tech companies to ensure that markets stay open and fair;
• Building infrastructure for storing and processing data: Between 2021-2027, the Commission will invest €2billion in a “High Impact Project on European data spaces and federated cloud infrastructures”. This project should increase the integration of European cloud infrastructures through the adoption of standards for data sharing, best practices, tools and governance mechanisms;
• Promoting “common European data spaces” in strategic sectors and domains of public interest such as manufacturing industries, mobility, health or finance. The Commission will propose measures to create pools of data available for reuse as well as technical tools and infrastructures to exploit them.
3. Key take-aways of the AI White Paper
The AI White Paper presents the Commission’s proposals to boost investments in AI and provides some food for thought on what an AI regulatory framework could encompass if needed.
To boost investments in AI, key actions would include:
• Revising the coordinated plan between Member States by end 2020; The objective is to attract over €20 billion of total investment in the EU per year in AI over the next decade;
• Providing support to leading universities and higher education institutes to attract the best professors and scientists and offer world-leading masters programmes in AI;
• Supporting the creation of “excellence and testing centres” for AI through the relevant EU funding programs (Digital Europe Programme, Horizon Europe);
• Setting up a new public private partnership in AI, data and robotics in the framework of the Horizon Europe programme;
• Preparing an action plan on the adoption of AI by the public sector.
The Commission will carry on the assessment of the regulatory framework to see whether any limited changes for identified problems/ risks are needed, with a special attention towards:
• The enforcement of existing EU and national legislation on fundamental rights (e.g. data protection, privacy, non-discrimination), consumer protection, and product safety and liability rules;
• Software embedded in products, to see when the EU product safety legislation and when specific rules such as the Medical Device Regulation apply;
• Safety risks, such as those concerning software updates for AI / Machine learning or the allocation of responsibilities between economic operators.
If a regulatory framework for AI is needed, then the Commission’s approach favors:
• A risk-based approach, with restrictions for ‘high-risk’ AI applications;
• A ‘high-risk’ AI application should meet two criteria:
o it is employed in a sector where significant risks can be expected to occur such as healthcare, transport, energy and several public sector activities (or in an activity considered high-risk such as recruitment processes or surveillance);
o the AI application in the sector in question is used in such a manner that significant risks are likely to arise, e.g. risks of injury, death or significant damage.
• Defining the types of requirements that ‘high-risk’ applications should meet. Such requirements could refer to: the types of data used to train the AI systems, the keeping of records and data, the provision of information on the AI systems to users, the robustness and accuracy of AI systems, the degree of human oversight, and the use of remote biometric identification;
• Voluntary labelling for no-high risk AI applications and conformity assessments for high-risk applications;
• Defining a governance framework that involves cooperation between national competent authorities and the consultation of a wide array of stakeholders.