‘AI First’ to ’Purpose First’: Rethinking Europe’s AI Strategy

Abstract

This paper examines the European Commission’s “AI First” strategy, arguing that it places acceleration and economic competitiveness above democratic values, societal benefit, and human-centric innovation. While substantial investment in AI is welcome when it promotes sustainable, equitable, and responsible innovation, the authors warn that policy is shifting from governance to unchecked deployment, risking fragmentation, dependency, and misaligned priorities. Rather than asking how AI can be applied, the paper urges policymakers to ask why, advocating a “People First” approach grounded in societal needs, digital sovereignty, and responsible innovation. The authors argue that Europe’s AI leadership should be shaped not by speed, but by principled direction, inclusivity, and a commitment to long-term public value.

 AI First

AI is increasingly being framed as a strategic imperative for economic growth, competitiveness and innovation. Yet, this purpose is often at odds with a more fundamental question: what should the purpose of AI be, and under which conditions does it genuinely add value to society? Following the recent launch of The European Commission’s (2025a,b) Apply AI Strategy and the ambitious InvestAI Programme, aimed at building pan-European AI “gigafactories”, (European Commission 2025c), heralded by the Commission’s President as a cornerstone for Europe’s AI competitiveness, the policy discourse has shifted from governance to acceleration. This rhetoric of Europe becoming the “Continent of AI’” however may signal a worrying departure from Europe’s longstanding commitment to human-centric and responsible innovation.

The timing and framing of the European Commission (2025a) “AI First” narrative appears to be closely aligned with the recommendations of the Draghi Report (European Commission 2025d), which emphasises digital investment and competitiveness as central to Europe’s economic renewal. While the substantial funding and incentives for AI research and innovation are welcome, the framing of “AI First” ignores a deeper set of concerns, including the limited evidence, if any, of substantial productivity and societal gains from AI use ((Estrada 2025; Wearden 2025). As such the shift to “AI First’” not only threatens to erode the foundations of Europe’s long-standing commitment to human-centric and rights-based innovation, leaving citizens, both in Europe and beyond, as the ultimate losers.

Full Steam Ahead, But What’s The Heading?

Despite Europe’s foundational focus on trustworthy and human-centric AI, recent Commission announcements, and public statements from its leadership, suggest a radical shift away from precaution, governance, and shared responsibility on AI, to a position of acceleration and competitiveness. AI is seen as a means to bolster economic growth through a highly ambitious industrial policy. What this perspective overlooks, both in Europe and globally, is a clear “people first’” perspective: recognition that technology must serve human and societal goals, not the other way around. The “AI First’” approach glosses over this vital point. While, lip service is paid to an assessment of the benefits and risks of the technology, these are framed as checks and balances, and fail to asks, for example, if a non-AI solution may be better or safer.

This acceleration approach also is in direct contravention of the explicit instructions of the EU Parliament (2024), which called for stronger precautionary measures, transparency, and accountability in the design and deployment of digital technologies to safeguard human rights, democratic oversight, and consumer protection within the EU single market. On the other hand, the EU has recently been on the receiving end of considerable criticism from key industry actors in Europe and beyond, who claim overregulation is killing competition, supposedly leaving industry vulnerable and driving skilled professionals to Silicon Valley (Haeck 2025). At the same time, concerns about a potential generative AI bubble burst have been raised by industry leaders and governments (Makortoff 2025), sowing fear of an economic collapse. The AI First policy can thus be understood as a response to mounting pressure to increase investment, reduce regulatory constraints, and accelerate AI deployment across society. In doing so, the European Commission has effectively adopted a full-steam-ahead approach to AI, yet without the coherence, governance frameworks, and people-centric orientation necessary to ensure that such acceleration aligns with Europe’s foundational values and long-term public interests. The European Commission must also recognise that framing AI development as a global race is both misguided and counterproductive, because such a narrative reduces a complex societal transformation to a contest of speed, rather than a question of direction, purpose, and public value. Moreover, Europe will not win any AI race. The US is too dominant in the currently popular massive, centralised approaches to AI, with the EU being too dependent on the US for the tech stack that allows the most pervasive forms of AI to function. AI leadership and digital sovereignty will not come from a fragmented approach where Europeans are told to see if and where AI can be wedged into sectors and society at large. Strategic leadership, a focus on key areas of innovation, how Europe’s limited resources can be used to maximise both economic growth and social good are key. An exploratory and human rights-driven alternative is more suitable and aligned with Europe’s values and aims than trying to keep pace with the US at any cost. An AI First policy will only further fragmentation, increase inefficiencies, undermine the EU’s competitive advantage and increase its dependency on non-European actors.

This is a pivotal moment to reflect not only on how we govern AI in the EU, but why we are developing and deploying it in the first place. Too often, we see technology placed before purpose, and innovation before inclusion. So, if not AI First, what is the right question? And how can poorly resourced actors, such as SMEs, civil society, universities, small EU countries and those in the global south with limited AI literacy make this assessment?

Not AI First, But AI Where It Is The Best Solution

Rather than presuming that AI, as claimed by the Commission’s President Ursula von den Leyen, will inevitably deliver “smarter, faster, and more affordable solutions’” (von der Leyen 2025), Europe must first determine where, and whether, AI genuinely serves societal needs.

That is, we must start with Question Zero: Why AI? (Lindstrom 2025). What problem are we trying to solve? Is AI truly the right or only solution for each case where it is being applied or promoted? Who benefits, and who bears the costs? By asking these simple questions, we quickly realise that sometimes, not always, AI is the answer. This approach offers a quick, low-cost way to assess AI’s relevance, especially useful for poorly resourced actors, who are often expected to adopt AI without sufficient AI literacy, resources, or support.

Europe As An AI Leader

AI is not inevitable, nor is its current trajectory predetermined. The EU has real choices to make. As such, the EU need to focus their efforts on actively navigating the correct path forward, rather than assuming that the choices have been made for them, and the only thing they can do is try to catch up. This ability to make choices is what digital sovereignty really means. Having the ability to decide over our futures. While the Commission’s suggestion of AI First may miss the mark, the EU retains the power to define when and how AI should be used, and, vitally, when it should not. By doing so, the EU can lead not through speed, but through purpose, setting a global example of responsible innovation that strengthens independence, upholds democratic values, and turns digital sovereignty into a shared regional strength.

References

Lindström, Anna Dahlgren, Virginia Dignum, Peter Ericson, Tatiana Titareva, and Jason Tucker. 2025. Responsible AI Self-assessment Workshop: Start with Question Zero. https://aipolicylab.se/2025/09/05/responsible-ai-self-assessment-workshop-start-with-question-zero/ AI Policy Lab, Published September 5, 2025. Accessed October 9, 2025.

European Commission. 2025a. Apply AI Strategy. Shaping Europe’s Digital Future. https://digital-strategy.ec.europa.eu/en/policies/apply-ai Accessed October 9, 2025.

 European Commission. 2025b. Communication from the commission to the European Parliament and the Council. Apply AI Strategy (Brussels, 8.10.2025, COM(2025) 723 final). https://ec.europa.eu/newsroom/dae/redirection/document/120429

European Commission. 2025c. EU launches InvestAI initiative to mobilise €200 billion of investment in artificial intelligence. https://digital-strategy.ec.europa.eu/en/news/eu-launches-investai-initiative-mobilise-eu200-billion-investment-artificial-intelligence Last update: 12 February 2025; Accessed October 9, 2025.

European Commission. 2025d. The future of European competitiveness. Part A: A competitiveness strategy for Europe. https://commission.europa.eu/document/download/97e481fd-2dc3-412d-be4c-f152a8232961_en?filename=ThefutureofEuropeancompetitiveness_AcompetitivenessstrategyforEurope.pdf European Parliament. 2024. Addictive design of online services and consumer protection in the EU single market (P9 TA(2023)0459). Official Journal of the European Union C/2024/4164 (2024). https://eur-lex.europa.eu/eli/C/2024/4164/oj/eng Accessed October 9, 2025.

Wearden, Graeme. 2025. Entry-level workers face AI ‘job-pocalypse’; US probes Tesla’s self-driving system – as it happened. The Guardian (2025). https://www.theguardian.com/business/live/2025/oct/09/water-customers-bill-hike-winter-blackouts-risk-falls-stock-markets-pound-ftse-business-live-news Retrieved October 9, 2025.

Makortoff, Kalyeena. 2025. Bank of England warns of growing risk that AI bubble could burst. The Guardian (2025). https://www.theguardian.com/business/2025/oct/08/bank-of-england-warns-of-growing-risk-that-ai-bubble-could-burst Accessed October 2025.

Haeck, Pieter. 2025. Dutch chips giant ASML executive Roger Dassen slams EU AI overregulation. https://www.politico.eu/article/dutch-chips-giant-asml-executive-roger-dassen-slams-eu-ai-overregulation/ Accessed October 9, 2025.

Estrada, Sergio. 2025. MIT Report: 95% of Generative AI Pilots at Companies Are Failing. Fortune (2025). https://fortune.com/2025/08/18/mit-report-95-percent-generative-ai-pilots-at-companies-failing-cfo/ Retrieved October 9, 2025.

Von der Leyen, Ursula. 2025. From now on it’s “AI First” — today the European Commission launches its new approach to artificial intelligence. LinkedIn post. Available at: https://www.linkedin.com/posts/ursula-von-der-leyen_from-now-on-its-ai-first-today-the-activity-7381720419516452864-2aUp.

Keywords (comma separated):

European Commission, European Union, Invest AI, Apply AI, AI First Policy, Question Zero, Responsible AI