With organisations increasingly adopting AI and overcoming some of the knottier implementation challenges, focus will continue to be on enterprise-wide implementation that can really deliver a competitive advantage. But realising this vision also requires tackling significant challenges: from upskilling employees, to overcoming scalability barriers to shift from experimentation to strategy.
Beneath the surface of AI’s promise lies a significant issue: scalability. While organisations can deploy small-scale AI pilots with relative ease, expanding AI across an enterprise remains a much harder challenge. Legacy systems, data silos and fragile AI governance structures are persistent barriers. But one factor looms particularly large: workforce readiness.
Upskilling employees to work alongside AI is therefore no longer optional. Currently, the UK AI skills gap exists on two levels: employees sometimes lack the basic fluency to use AI tools effectively, while organisations face a broader shortage of data scientists, engineers and other technical talent. Bridging these gaps is essential to scaling AI successfully.
From experimentation to enterprise impact
Too many organisations remain stuck in the experimental phase of AI. They test the technology in isolated use cases, such as deploying chatbots or enhancing a single operational process, but fail to scale it to its full potential.
Data supports this view. According to McKinsey, 88% of organisations now use AI in at least one function, up from 78% last year. This is a seismic leap in adoption, showing AI is no longer experimental; it’s operational. However, most companies are still caught in an experimentation and pilot mode, with just one-third reporting that they’re scaling AI across the enterprise.
To accelerate adoption, organisations must focus on clearly measurable business outcomes, as opposed to isolated experiments without a framework. Integration should be guided by specific goals, such as boosting customer retention, reducing operational costs or accelerating product development cycles.
Transforming customer experience
Hyper-personalisation is already reshaping how businesses interact with customers. But by 2026, hyper-personalisation will evolve from being a competitive differentiator to an expectation among customers.
Industries such as insurance, retail and financial services are making headway in demonstrating its potential. Across these sectors, AI is enabling real-time predictive risk assessments and even preventative advisory services – all personalised to the customer. Speaking at a recent industry panel, Natasha Davydova, CIO of AXA UK&I, noted how the insurance sector, specifically, is leveraging AI to streamline claims processes and provide proactive support. In one case, she described a future where claims for delayed flights could be processed automatically, without the customer needing to lift a finger.
Looking ahead, 2026 will not be without obstacles. Organisations must focus on securing consumer trust through better governance, ensuring transparency in how they use data and clarifying AI’s role in decision-making processes. Leadership will also play a pivotal role by driving cultural shifts within organisations and fostering collaboration across departments.
Sebastian Weir is the Executive Partner of AI, Analytics & Automation Practice Leader at IBM
Read more: The UK government’s digital ID project costing £1.8bn isn’t the real scandal
