A new AWS report found AI adoption is accelerating rapidly in the U.K., with nearly two-thirds of organizations now using the tech. Yet while adoption is widespread, transformation through AI is uneven, and only a small percentage have reached the advanced stage (where AI forms part of core business processes and decision-making).
In this Q&A, Alison Kay, vice president and managing director for the U.K. and Ireland at AWS, discusses the report’s findings and how businesses in the U.K. (and beyond) can adapt to make the most of AI.
What stood out to you as the key findings from AWS’s latest AI adoption report?
Alison Kay: Businesses are no longer asking whether AI matters; they are asking how to use it to grow, compete, and innovate faster.
But while adoption is surging, the share of organizations using AI’s advanced functionality has risen by just one percentage point in the past year. Most organizations are still using AI for basic tasks like drafting emails, summarizing documents, and running off-the-shelf chatbots. This is like buying a smartphone and only using it to make phone calls. Just 24% of U.K. adopters have reached the advanced stage, where AI is genuinely integrated into core business processes and decision-making.
This gap between basic and advanced AI adoption matters because the economic value of AI scales dramatically with depth of use. Closing it could unlock an estimated £35 billion ($46.8 million) in unrealized productivity gains by 2030 — roughly the equivalent of the entire annual economic output of the city of Manchester. This would represent a significant contribution to reversing the productivity slowdown that has held the U.K. economy back for over a decade.
You’ve described AI as being at a pivotal moment in the U.K. Why does this feel like such a significant time?
Kay: The pace of technological advancement is accelerating. The leap from generative AI to agentic AI — from reactive assistants to proactive, autonomous systems that can understand, decide, and act with minimal oversight — happened in months. Eighteen months ago, most people weren’t even talking about agentic AI.
These advances are compressing innovation cycles, shortening the time it takes for new ideas to move from research to commercial application and widespread deployment. Ultimately, firms that adopt advanced AI can move from idea to product to scale faster than those still relying on basic tools. The risk is that differences in capability and performance widen over time — not only within the U.K., but globally as well.
However, most U.K. organizations aren’t prepared for what’s already here. Only 22% have heard of agentic AI, and, of those, just 4% have deployed it. Meanwhile, only 21% said they feel ready for advanced AI. This matters because the U.K.’s long-term opportunity will depend on how quickly organizations scale beyond these initial use cases.
Closing this gap will require action on three fronts: driving AI adoption through stronger incentives and support for organizations to move from experimentation to transformation, scaling AI across public sector services, and building workforce skills at pace.
What barriers are preventing organizations from moving beyond experimentation, and how can they be overcome?
Kay: Half of organizations cite AI and digital skills shortages as the biggest challenge to expanding their use of AI. The faster organizations adopt AI, the more acutely they feel the absence of the skills needed to use it effectively. Only 17% of organizations say they have a strong AI skillset today, yet 84% expect AI skills to be important in the next five years.
The U.K. needs to close the digital skills gap at pace by investing in training, public-private partnerships, and AI literacy at every level of education.
This year’s data also highlighted the role that government can play in increasing the U.K.’s AI adoption. More than three-quarters of organizations say they are more likely to adopt AI if the public sector integrates it into its own systems.
How is AWS working with government and industry to support AI skills development, and what does meaningful AI literacy look like in practice?
Kay: Two-thirds of U.K. employees said they want to learn new AI skills, but many don’t know where to start, and training provision remains uneven. Among those who have received training, the most common routes are informal or self-directed learning, suggesting a lack of structured pathways to build AI capability at scale.
That’s where collaboration between government, industry players, and education matters. We’re looking to close that gap.
Looking ahead, what will separate organizations that succeed with AI from those that struggle?
Kay: The organizations seeing the biggest gains are doing more than adding AI on top of existing workflows; they’re using it to fundamentally change their operating models.
To be successful with AI, organizations need to have a proper AI strategy, not just a pilot. Secondly, organizations need to invest in the right people: data scientists, machine learning engineers and AI product managers. Thirdly, they need to have modern infrastructure, typically cloud, which lets them build, test, and iterate without massive upfront costs. Those three things together create a compounding advantage. Get one right, and the others get easier.
Editor’s note: This interview has been edited for clarity and conciseness
