The growing digital divide
As Europe's AI revolution accelerates, a concerning pattern emerges. While overall adoption numbers paint an encouraging picture, they mask a deepening divide that could shape Europe's economic future for decades to come.
Only one out of every ten (13%) large businesses are using AI to deliver a new product or service for customers, while startups are driving change in this area (37%)
Startups are far more optimistic than large businesses - 25% more startups see launching a new product in the next three years as a realistic goal using AI if their business had the right capabilities
Startups report that AI plays a central part in their business operations, with 38% reporting that their business proposition is based around AI or that they are heavily reliant on it for their core service
The innovation gap
Picture two European businesses: a three-year-old startup in Stockholm and a century-old manufacturer in Lyon. Both use AI, both would count towards Europe's digital targets, but the similarity ends there. The Digital Decade targets do not capture the nuances between the two; thus, they risk giving Europe a false sense of security from rising adoption figures.
The startup has embedded AI into its very DNA - using it to create new products, reinvent customer experiences, and pioneer entirely new business models. Meanwhile, the manufacturer, despite its rich history and deep expertise, uses AI primarily for basic automation and efficiency gains.
This isn't an isolated example. Our research reveals that while 53% of large businesses are consistently using AI, only 14% have implemented the most advanced operations.
The contrast does not end there:
Beyond simple adoption
The divide goes deeper than just adoption rates -
it's about the very ambition of how AI is used. We've identified three distinct stages of AI maturity:
Basic automation of existing processes:
In the past year, 58% of businesses have undertaken AI to improve internal processes / existing processes using AI. This rises to 63% among large businesses.
Customer service chatbots:
In the past year, 56% of businesses have undertaken AI to implement new AI-based customer service solutions, such as chatbots.
Simple data analysis
In the past year, 54% of businesses (and 52% of large businesses) have been using AI to analyse large datasets.
Stage 1: First steps
Most large enterprises (73%) currently focus their use of AI on improving efficiency or productivity, i.e. the early steps of adoption for incremental gains, compared to 56% of startups.
These businesses are primarily using AI for:
While valuable, these applications only scratch the surface of AI's potential. In order to truly harness the unique potential of the technology, large businesses must consider how they can use the technology to truly transform their business and drive innovation, including looking toward the vibrant ecosystem partnerships with technology providers and government incentives — which are currently driving adoption.
Stage 2: Transformation begins
This is where businesses start to scale and deploy uses of AI across business functions.
Currently, only 12% of large enterprises are in this phase, compared to 17% of startups.
At this stage, businesses can:
Use AI to automate workflows: 47% of adopters report that they are harnessing AI for this purpose.
Reimagine customer experiences: 18% of businesses that are using AI report that this is their most advanced AI process.
Creating entirely new business models:
35% believe that AI will completely transform the industry through the creation of new business models in the process.
Fully integrating AI:
Crucially, only 3% of large enterprises have fully integrated AI in their processes and in the centre of their strategies for growth and innovation, compared to 18% of startups.
Developing their own custom AI solutions:
Only 4% of businesses that are using AI say this is their most advanced AI process.
Stage 3: Strategic reinvention
This is where the magic is happening for 14% of large businesses and 26% of startups. Businesses, at this stage, harness the most advanced AI systems that combine multiple types of AI tools or models to perform complex tasks.
At this stage, they are:
Despite growing AI adoption, most businesses remain in the early integration stages. While European startups are pioneering growth and innovation, large businesses risk falling behind and remaining in the early stages of adoption. The majority of businesses (61%) remain at Stage 1, a figure which rises to 73% among large established businesses. 26% of startups operate at Stage 3, compared to only 14% of large businesses.
This critical divide is a competitive disadvantage for individual large companies, and a threat to Europe’s economic future. While large businesses, which generate over half of the turnover in Europe,[1] continue to lag in digital adoption, Europe misses a significant opportunity to build on its existing leadership across industries, undermining its competitive position.
Investment patterns reflect the divide
This two-tier pattern is clearly visible in AI investment strategies:
Currently, 72% of startups report having a dedicated IT AI budget, compared to just 58% of large businesses, highlighting a clear distinction in resource allocation.
Furthermore, large businesses allocate only 15% of their IT budget to their AI development on average, a figure that rises to 20% among startups. This indicates a gap in priorities, where Europe's largest companies need to increase R&D investment to match startups' more intensive adoption of AI.
This investment gap risks widening the innovation divide.
This distribution highlights the untapped potential of AI, with significant room for growth as businesses progress toward more transformative stages of adoption and look to integrate the technology more deeply into their operations.
Despite sustained rapid growth in AI adoption, a two-tier AI economy is emerging, with many larger businesses remaining focused on the first stage of adoption, covering basic data analysis and driving efficiencies. By progressing through all stages of AI integration, from basic efficiency to strategic reinvention, businesses are able to move beyond surface-level benefits and become leaders in innovation. Prioritising all three stages to achieve the deepest level of integration enables long-term business growth, drives revenue, and ensures businesses remain competitive in a rapidly evolving, digitalised global economy. With cloud-enabled AI having the potential to add nearly $434 billion to the European economy - these ambitions toward full integration are pivotal.[2]
The HALO Trust is speeding up clearing landmines in conflict areas to save lives and empower local communities.
The non-governmental organisation is using AWS cloud infrastructure to analyse thousands of drone images of minefields securely using AI, significantly reducing analysis time from days to hours. This allows HALO to work more efficiently in heavily mined areas around the world, including Ukraine.
The HALO Trust features in AWS Pioneers Project: a spotlight on the pioneers behind Europe’s groundbreaking innovations shaping the future of AI.
2. Research from AWS and Telecom Advisory Services: https://www.aboutamazon.com/news/aws/ai-cloud-adoption-economic-impact-gdp-aws