Businesses face regulatory fragmentation, skills gaps, and limited financing
While interest in AI is high and adoption is accelerating, challenges threaten competitiveness. As outlined above, businesses face regulatory fragmentation, skills gaps, and limited financing.
Addressing Europe’s AI opportunity demands tackling these underlying capabilities and the constraints that determine whether businesses can embed AI effectively, scale across borders, and compete globally:
1. Regulatory fragmentation and complexity
Europe's fragmented regulatory landscape constrains AI scaling. Businesses cite key challenges: fragmented regulatory requirements across countries (41%), limited support for cross-border testing and deployment (36%), and market fragmentation hindering customer access (33%). According to IMF analysis, remaining internal challenges within the Single Market, including regulatory and procedural frictions, are equivalent to a 110% tariff on services, demonstrating how high these costs can be within the EU itself.
This suggests that Europe’s simplification agenda is not yet translating into reality for businesses. Despite efforts to streamline rules and reduce burdens, many organisations still describe a landscape defined by divergence and uncertainty, where scaling across Europe feels closer to expanding into multiple separate markets than operating within one. As highlighted by Letta in his 2024 report, and reiterated in his contribution to this report, fragmentation of the Single Market and asymmetry between territories and legal and tax systems end up increasing difficulties and multiplying obstacles to productive activity.
Fragmentation is both a policy issue and a practical challenge that directly affects speed, cost, and competitiveness.
This scaling friction compounds the broader compliance burden businesses already face as they navigate a lack of regulatory clarity across Europe:
To support their business’s compliance and regulation efforts, 48% currently have a dedicated team or function whose primary responsibility is AI governance and regulatory matters. This falls dramatically among startups, where just 9% have a dedicated compliance team.
Businesses estimate that 42% of their total tech spend goes towards compliance with national and international regulations, up from 40% last year. When asked what makes up this 42%, 50% of businesses cite relationship management (e.g. keeping clear documentation, communicating with regulators, filling out forms and filing queries) with the responsible government authorities, followed by legal consultations or external advisory services (44%), and employee training on compliance requirements (38%).
These burdens are felt equally regardless of company size. For SMEs and mid-sized businesses, the percentage was 42% and 43%, respectively, while for large enterprises, this figure stands at 41%. This figure is highest among startups, at 44%.
81% say compliance costs have increased over the past three years, and 80% expect these costs to rise further over the next three.
The urgency is real because competitors aren’t waiting. Other regions are moving faster, investing more aggressively, and making it easier for companies to scale. Emerging tech hubs, such as Abu Dhabi’s, are attracting some of the world’s most exciting tech startups, offering high levels of funding and taking steps to boost investment and create new jobs. Singapore has announced a landmark investment of over SGD 1 billion into its National AI Research And Innovation Plan, intending to harness AI for economic growth, social good, and global competitiveness. U.S. productivity has grown more than seven times faster than in the Euro area since 2019. Every year Europe delays simplification is a year its businesses fall further behind.
2. The digital skills gap
Skills shortages are the primary challenge to wider AI adoption in Europe, which can be overcome by concerted efforts by governments and businesses. Echoing findings from previous years, businesses do not lack interest in AI, but many lack the capacity to adopt it, whether due to shortages in AI-specific talent, limited workforce readiness, or insufficient internal budgets to invest in training with confidence.
Over half (51%) of businesses report that a shortage of AI and digital skills and a lack of current internal workforce capacity are preventing adoption or expansion of AI use, while three quarters (75%) of businesses report that their AI skillset requires improvement.
Businesses’ self-assessments show a gap between current capability and expected future need. Only a minority report having a strong AI skillset today, yet expect these skills to be increasingly important:
"There is a real urgency to this moment. We must succeed, and we must move quickly. But responsibility does not rest with European leaders alone. It also lies with all us – European citizens and citizens of the world – who want to see the European Union become more integrated and fulfil the missions outlined in this report."
Enrico Letta
Dean of the School of Politics, Economics and Global Affairs at IE University and Former Prime Minister of Italy
Hiring data reinforces the scale of the challenge. 35% of businesses report difficulty attracting local talent with the required digital skills and, on average, businesses estimate it takes 6.8 months to find an employee with the right digital skills once a role is posted. AI skills are increasingly seen as a baseline requirement across roles, not only specialist positions.
The skills gap is already constraining Europe’s competitiveness: 44% of businesses say shortages of digital and tech talent are holding Europe back. Without sufficient capability to implement, govern, and scale AI, firms risk falling further behind as adoption accelerates, turning skills shortages into an economic and growth constraint.
Businesses expect AI skills to be important
81% of businesses expect AI skills to be important in their industry over the next five years. Businesses estimate that AI literacy will be important for over half (53%) of jobs.
Training has stalled
Citizen training has stalled – only 33% report having undergone digital training.
Of these, 67% used online courses, followed by self-study using online resources (39%), and on-the-job training (36%).
Reported hurdles to training include cost (44%), a lack of time (34%) and lack of information about the training available (24%).
25% cite unclear ROI or an unclear business case as a hurdle. Uncertainty around benefits or use cases continues to delay investment.
However, among those businesses that have adopted AI, 56% report matching or exceeding their investment.
28% cite insufficient internal financial resources as a challenge to adopting or expanding AI and 43% say they do not have a dedicated AI budget.
30% say other business priorities take precedence over AI investment.
Access to finance remains a key constraint on AI adoption and scaling. While 89% of businesses expect AI to account for a larger share of their IT spend over the coming years, many remain cautious in the near-term. They often lack dedicated AI budgets or allocate only a small share of IT spend to AI.
3. Access to funding and investment
56% say government support – such as grants and tax incentives – is crucial or very important in their decision to adopt AI, down from 67% last year.
%20also%20report%20a%20lack%20of%20incentive%20or%20external%20support%20to%20innovate.png)
A fifth (20%) also report a lack of incentive or external support to innovate.
Public sector demand could support wider diffusion, but procurement processes remain challenging:
34% of businesses cite complex or slow public procurement processes as a challenge to scaling AI solutions.
30% say opportunities to sell to the government are crucial or very important in their decision to adopt AI.
Limited support and incentives risk reinforcing a capability gap. Businesses with sufficient capital and skills are progressing faster, while others struggle to build the capacity required for transformation. Firms allocating a higher proportion of their budgets to AI are reporting faster innovation cycles and higher ROI. Without stronger incentives, particularly those that reduce risk and build confidence in investment, AI adoption may remain concentrated in experimentation rather than widespread transformation.
In social care, time is the most valuable resource. CareMates is Germany’s first AI-powered software for patient admissions, cutting admission time from five hours to one. Through digital forms, automated workflows and AI-generated documentation built on AWS infrastructure – including Amazon Bedrock for access to multiple foundation models – CareMates is able to deploy secure, privacy-compliant AI rapidly across hundreds of facilities, without having to manage its own infrastructure. CareMates is streamlining admissions at scale, returning thousands of hours to care professionals and directly addressing workforce shortages in one of Europe’s most stretched sectors.
CareMates reflects Europe’s strength in mission-driven entrepreneurship and illustrates how AI adoption can move beyond experimentation into operational transformation when there is clear, streamlined access to support and funding pathways for Europe’s most innovative startups.
"For an early-stage company in healthcare, speed and trust are everything. Building on AWS meant we could launch quickly, meet data protection requirements, and focus our capital on improving our product, not managing infrastructure. With clearer funding pathways and simpler scaling conditions, startups like ours can deliver impact across Europe much faster.”
Johannes Kiwi
Technical Lead and co-founder
CareMates
Europe’s AI landscape is at a crossroads. The continent is producing world-class founders, research, and early-stage innovation at an unprecedented pace. Yet as this report makes clear, the challenge Europe now faces is not one of ideas or ambition, but of commitment and scale.
At Atomico, we work with founders across Europe who are building some of the most advanced AI companies in the world. What we see consistently is that Europe excels at creation but struggles with continuation: moving companies from early success to global leadership. This report makes clear that funding pathways are central to this challenge. While early-stage capital is increasingly available, the routes to sustained growth – such as late-stage funding and deep pools of scale capital – remain fragmented and uncertain.
This matters more than ever as AI innovation cycles compress. Companies building with advanced, agentic AI can iterate, deploy, and expand at unprecedented speed. In this environment, delays in funding, regulatory uncertainty, or barriers to cross-border scaling translate directly into lost momentum. For many of Europe’s most promising startups, the decision to scale abroad is not ideological – it is pragmatic.
Europe’s opportunity is significant. The continent has the talent, technical depth, and entrepreneurial energy to be the best place in the world to start, build and scale a company. But to realise that potential, Europe must strengthen the conditions that allow companies to grow where they are founded: clearer funding pathways, deeper capital markets, and a regulatory environment that rewards ambition rather than caution alone.
The future of European competitiveness will be shaped by whether its best companies can scale at home and whether Europe can match its strength in innovation with equal strength in growth.
Tom Wehmeier
Partner and Head of Intelligence
Atomico