The AI Paradox: Mid-Market Confidence Masks a Crisis of Capability

LONDON – July 08, 2026 – A striking paradox is unfolding within the engine room of the British and Irish economies. While mid-market companies express overwhelming confidence in their ability to harness artificial intelligence, a new report reveals that the vast majority are failing to move AI projects beyond the pilot stage, stalled by a critical lack of genuine expertise, poor data foundations, and absent governance.

The research, published by specialist AI consultancy Klarus, surveyed 500 senior decision-makers at mid-market firms—those with revenues between £200 million and £2 billion. The findings paint a picture of ambition clashing with reality. While nearly three-quarters (73%) of these companies have begun deploying AI, a mere 10% have successfully scaled all their initiatives. This disconnect highlights a crucial implementation gap that threatens to stifle innovation and productivity in a sector vital for national economic growth.

A Crisis of Overconfidence

At the heart of the problem is a significant disparity between perceived and actual capability. According to the Klarus report, "The state of AI in the mid-market," an average of 91% of companies are confident in their internal expertise across all facets of AI deployment. Yet, in a telling contradiction, "lack of AI expertise" was cited as the joint top reason (48%) for AI projects failing to progress.

This suggests a widespread Dunning-Kruger effect in the corporate world, where initial enthusiasm for generative AI tools and accessible platforms is mistaken for the deep, specialized knowledge required for enterprise-grade implementation. The remaining 90% of companies that have explored AI are left with a portfolio of stalled pilots or initiatives stuck in the early stages of development, unable to translate initial promise into tangible business value.

"Mid-market companies have a real advantage because they can often move fast, particularly when it comes to technology transformation," said Alper Gunaydin, CTO at Klarus. "However, our research shows that too many pilots stall because companies lack AI expertise, quality data and effective governance. Making that agility count requires clear priorities, strong foundations and access to senior expertise, all of which will help translate investment into tangible business outcomes and unlock growth without increasing the cost base.” The findings suggest an awakening is underway, as 39% of respondents now list building internal expertise as a top priority for the next year, indicating a shift from blind confidence to a more sober focus on execution.

The Cracks in the Foundation: Data and Governance

Beyond the skills gap, the research exposes fundamental weaknesses in the technological and procedural bedrock necessary for AI to thrive. An overwhelming 83% of mid-market firms that have piloted or deployed AI report experiencing poor data quality, with 69% stating it is actively preventing or delaying their AI activities. AI models are only as reliable as the data they are trained on; without clean, structured, and relevant data, any AI initiative is built on sand.

This data dilemma is compounded by a severe lack of formal oversight. More than half of the companies surveyed (59%) have yet to establish a comprehensive AI governance framework, meaning they operate without formal policies, controls, or both. This absence of guardrails was, alongside lack of expertise, the other top reason for stalled pilots, cited by 48% of respondents. Concerns over ethics, security, and privacy paralyze projects, as companies are unwilling—and rightly so—to risk deploying powerful but unchecked technology into their core operations.

The path to success, however, is clearly illuminated by the few who have scaled their projects. For those whose AI initiatives met or exceeded expectations, strong data quality was the single most critical success factor (cited by 59%), followed closely by the presence of effective governance, security, and ethical controls (54%). Improving data quality is now the joint top priority for the coming 12 months (43%), while 35% plan to finally strengthen their AI guardrails, signaling that the mid-market is beginning to understand that the unglamorous work of building solid foundations is non-negotiable.

A Stalled Engine for Economic Growth

The failure of the mid-market to effectively scale AI is more than just a series of individual corporate disappointments; it represents a significant drag on national economic ambitions. In the UK, these firms constitute just 0.5% of companies yet generate an outsized 30% of the country's Gross Value Added (GVA), as noted in a 2024 NatWest report. The Irish mid-market is similarly crucial and deeply intertwined, with a recent Enterprise Ireland survey showing 64% of Irish firms maintain a permanent physical footprint in the UK.

When this economic engine sputters on a key technological transformation, the repercussions are widespread. "The findings come as AI adoption becomes a central priority, with UK government policy now focused on closing the productivity gap by embedding AI across businesses," commented Tim Flagg, CEO of UKAI, the UK's trade association for AI businesses. "While tech giants often dominate headlines, it is the mid-market that will be the true driver of this transition."

Flagg's perspective frames the Klarus findings as a national call to action. The common hurdles of establishing guardrails, building capabilities, and implementing robust governance are not just internal business problems but systemic challenges that must be addressed collaboratively between industry and government. For the UK and Ireland to realize their AI-powered economic goals, it is this "critical middle" that must successfully turn policy into mainstream economic reality.

Beyond Automation: Reshaping the Workforce

While the report highlights significant challenges, it also offers a more nuanced and optimistic view of AI's impact on the workforce, moving the conversation beyond simplistic fears of automation and job loss. When asked about the primary impact of AI on junior staff, nearly half (45%) of respondents said it is enabling them to do their jobs better or quicker. Furthermore, a quarter (24%) reported that AI is actively creating new roles and opportunities within their organizations.

This suggests that, for many, AI is functioning as a powerful augmentation tool, automating mundane tasks and freeing up human employees to focus on higher-value work, critical thinking, and creative problem-solving. It positions AI not as a replacement for human talent, but as a catalyst for workforce evolution. This transformation allows junior employees to contribute more significantly and earlier in their careers, while creating demand for new skills in AI management, data science, and ethics. This trend underscores the importance of strategic investment in reskilling and upskilling programs to prepare the workforce for an AI-enabled future, turning a technological shift into a human capital advantage.