The new president of the European Commission, Ursula von der Leyen, has recently unveiled her policy agenda for the next five years. As part of her proposal to create “a Europe fit for the digital age” she has promised to put forward legislation “for a coordinated European approach on the human and ethical implications of AI” in the first 100 days. This commitment has ensconced AI as a top priority for policymakers under the next Commission. This is a welcome direction: The EU needs to reform its AI policy if it hopes to compete with the United States and China in the global race. But a key challenge for the new president will be to lead EU policymakers away from Europe’s typical “regulate first” approach and towards a focus on growing the investment, data, and talent required to develop AI and accelerate its adoption, and creating an innovation-friendly regulatory environment across the EU.
The next Commission should thus focus on three main priorities.
First, EU policymakers should prioritize completing Europe’s digital single market initiative—an objective that the former Commission has long promised but has not delivered. Innovative U.S. companies have quite an advantage over their European counterparts because its large market for digital products and services allows U.S. companies to scale up, whereas Europe’s digital market remains highly fragmented.
Second, EU policymakers should foster the building blocks of AI, namely R&D, investments, and talent. By accelerating the digitalization of the public sector in Europe, and by increasing and opening up their troves of data, EU governments will be able to provide European businesses and research hubs with the ingredients required to develop European AI. In her announcement, the Commission’s newly appointed president vowed to “prioritize investments in AI” through public-private partnerships, “get Europe up to speed on digital skills” by updating existing education plans, and rightly mentioned the “important role” of the public sector in “stimulating digital transformation.” With the next update of its coordinated action plan on AI planned for the end of this year, the EU has an opportunity to consolidate EU member states’ national AI strategies, to make these promises a reality.
Third, EU policymakers should modify existing regulations such as the General Data Protection Regulation (GDPR) and resist calls to transform ethical guidelines into requirements for the development of European AI. Indeed, concerns over the negative impact of the GDPR are mounting within Europe’s business community, particularly among startups. The risk of the Commission’s president pledge is that it will lead to a regulatory framework designed to slow and constrain, rather than accelerate AI. For example, EU policymakers could very well expand when organizations must provide explanations for decisions made by automated systems, which the GDPR already requires, by applying this requirement to situations where no personal data is involved. Such a requirement is impractical for AI, as there is a trade-off between accuracy and transparency in algorithms. If EU policymakers want to achieve “better regulation,” they should propose sector-specific rules, and combine these with soft law instruments such as codes of conduct, in consultation with digital stakeholders.
The next Commission president is right to focus on AI, but given the mistakes the European Commission has made with data protection, many are justifiably concerned about their intentions. EU policymakers can allay many of those concerns by outlining a roadmap for supporting the development of AI in Europe by strengthening the digital single market, supporting AI R&D, investment, and talent, and rethinking its approach to digital regulation.
Image credit: Flickr user European Parliament