The UK is positioning itself as a global hub for AI innovation, signalling serious intent to lead in both capability and scale. Three companies recently committed to £14 billion investment to build the AI infrastructure required to harness the potential of this technology and deliver 13,250 jobs across the country. That’s on top of the £25 billion in AI investment announced by the government at the International Investment Summit.
This wave of investment reflects a wider shift in mindset. AI is no longer seen as a side experiment or distant prospect. It’s moving to the centre of business strategy. In customer engagement especially, it’s gaining traction fast. Leaders are no longer debating whether to invest, but how quickly they can scale and integrate AI to stay competitive.
Increasingly, the push for AI adoption is coming from the top, with CEOs and boards approving significant investments often before clear use cases are fully mapped out. But early ambition doesn’t always translate into lasting impact. A recent IBM study found that only 25% of AI initiatives have delivered the expected return, and even fewer are successfully scaled.
This gap between intent and impact matters more than ever. As customer interaction becomes increasingly digital, and available 24/7, the ability to deploy AI effectively will define who leads. However, what’s less frequently discussed is the role that consumption models play in turning early AI ambition into lasting impact.
The infrastructure and funding challenge
Most legacy infrastructure wasn’t built to support the demands of AI. Whether that’s scalability, continuous innovation, or the agility modern use cases require. At the same time, organisations are wrestling with how to fund AI initiatives as they take up an increasing share of budgets. Genesys research shows that leaders expect over a third (33%) of their customer experience budget to be spent on AI in the next year. Yet despite this projected increase, only 1% of companies believe their AI investments have reached maturity, according to McKinsey. The disconnect between aspiration and reality is clear.
Part of the problem lies in how AI is priced and delivered. Current models, which range from license and subscription-based to consumption, freemium, and outcome-driven, often fall short of what businesses really need. Many want the freedom to start small, test new ideas, and scale on their terms. But rigid pricing structures can lock them into long-term commitments or static capabilities, while more flexible options can introduce unpredictable costs. Even outcome-based models, though promising, can add complexity when it comes to tracking performance and accountability.
Still, pricing alone isn’t the core issue. Many organisations struggle because they lack a strategy for scaling AI effectively across the business. In customer experience, that means going beyond pilots to fully integrating solutions like virtual agents, predictive routing, copilots and automated workflows -connected in a way that drives continuous learning and value. These tools only deliver their full potential when backed by a flexible platform and commercial model that supports rapid iteration. Without that, many businesses delay AI investments simply because they can’t forecast value clearly or adapt fast enough. The path forward requires not just funding but demands a foundation built for change.
Token-based models: A strategic advantage
As more organizations begin to deploy agentic AI to fuel always-on business, the right consumption model could be the difference between failure and a competitive edge by enabling rapid experimentation and scaling without financial risk. According to a recent survey from Capgemini Research Institute, more than 50% of respondents say they favour a consumption-based pricing model for agentic AI solutions. One approach that achieves this is token-based models. They provide businesses with a currency-like approach to AI usage where resources can be allocated to specific outcomes, such as virtual agents, AI-driven conversation summaries, or autonomous workflows.
As AI-driven customer interactions become more widespread, this approach will be increasingly important. Tokens act as the currency that supports scalable, value-aligned AI usage without adding complexity. This ensures businesses can manage high volumes of AI activity around the clock while keeping costs under control.
Token models also promote experimentation. Organisations can start with pay-as-you-go tokens to explore capabilities and respond to business needs without lengthy contracts. For example, they might boost digital self-service in peak periods or introduce auto summarisation to improve agent efficiency. As usage grows, they can move to committed token bundles for greater predictability, while still retaining flexibility to reallocate tokens across different use cases.
Choosing the right economic model for AI success
With AI becoming more embedded in business operations, often working alongside or even ahead of human teams, choosing the right economic model is crucial. It can be the difference between simple experimentation and driving impact across the entire enterprise.
Leaders need to consider whether their AI consumption is built for continuous, autonomous activity rather than just supporting human-led tasks. They should also evaluate if their AI platform allows the flexibility to evolve use cases without the burden of renegotiating contracts or committing excessive resources. Additionally, transparency, predictability, and adaptability in pricing are essential factors to ensure the economic model aligns with business needs.
Admiral exemplifies an organisation that has identified the right modelling structure to align with its customer experience strategy. When the company first set out to improve its customer experience operations, AI was a key part of that vision. Since modernising its infrastructure with Genesys Cloud, Admiral has empowered nearly 500 agents with copilots and other AI tools including workforce engagement management, providing critical insights into employee performance and business opportunities.
The next step is to accelerate the AI roadmap with predictive engagement, which will allow Admiral to personalise interactions and drive strategic business growth priorities. The auto insurer is using the AI token model from Genesys to rapidly test and innovate, scale at its own pace, and focus on developing the AI capabilities that deliver the greatest business value.
Thinking wisely for the future
The UK’s ambitious AI investments and shifting leadership mindsets signal a clear commitment to harnessing this transformative technology. However, as organisations grapple with integrating AI into their customer experience strategies, they need consumption models that support their innovation ambitions rather than constraining them with rigid contracts and unpredictable costs. The gap between early AI investment and lasting impact will be bridged by those who choose models that are most suitable for their needs.
In the experience economy driven by AI, success won’t just come from what businesses deploy – it will come from how wisely they consume it.
Olivier Jouve
Olivier Jouve is Chief Product Officer at Genesys. where he leads the product, artificial intelligence, and digital teams. Before stepping into this role in 2022, he served as Executive Vice President and General Manager of Genesys Cloud and Head of AI development. Prior to joining Genesys, Olivier held multiple senior executive roles at IBM, including Vice President of Offering Management for IBM Watson IoT.
