New Research Reveals Challenges in Enterprise AI Adoption
London, November 17, 2025 – New research from the diginomica network uncovers the challenges enterprise organizations face in adopting artificial intelligence (AI). Based on in-depth discussions with 35 CIOs and CTOs from major global companies, the report highlights a significant disconnect between expectations and reality regarding AI’s return on investment.
According to the research, 93% of diginomica network members have implemented AI within their organizations. Use cases range from website chatbots and predictive algorithms to drug discovery. Despite this widespread adoption, however, high-profile AI implementations—such as Microsoft Copilot and automated bid tools—have failed to meet the elevated expectations set by boards and executive teams.
Key Findings
- Over three-quarters of organizations report AI success rates exceeding 50%.
- Technology leaders consistently emphasize that AI has not met the expectations placed upon it by boards and executives, making it difficult for CIOs to manage stakeholder expectations and maintain trust.
- Data quality and change management, rather than technology itself, are the primary barriers to realizing AI’s full potential.
- Poor adoption strategies are causing organizations to capture as little as 10% of potential AI benefits.
Ian Cohen, a seasoned CIO with leadership experience at Lloyds, the Financial Times, and Addison Lee, stresses the need for CIOs to manage the growing hype around AI. He explains, “It is our job to temper that enthusiasm with a healthy dose of reality.”
Barriers to Successful AI Adoption
The report identifies several critical challenges hindering AI success:
- Poor data quality
- Insufficient change management
- Legacy system constraints
- Fundamental misunderstandings of AI capabilities across the C-suite
Confusion around different AI technologies—including agentic AI, generative AI, and robotics—is also highlighted in the research.
The study emphasizes the importance of implementing proper change management strategies and warns against treating this aspect as an afterthought. Without effective adoption strategies, organizations risk repeating the failures of previous technology implementations.
Cohen adds, “We risk the same happening with AI if we do not properly drive education and adoption.”
About the Research and diginomica Network
Mark Chillingworth, author of the research and facilitator of the diginomica network, thanks the participating CIOs and CTOs for sharing their experiences and valuable insights.
The full report, “CIOs navigate AI’s weight of expectation and explore opportunities,” is available for download here.
About diginomica
diginomica is an independent media and analyst organization serving enterprise leaders in the digital era. With experienced writers and analysts based in the US and Europe, diginomica delivers authoritative reporting and analysis that cuts through the noise.
Unlike traditional advertising-driven media, diginomica prioritizes reader quality, engagement, and influence. The organization provides rich insights based on daily interactions with both technology buyers and vendors. diginomica is also committed to the responsible use of generative AI in journalism, focusing on reader experience, accessibility, and trust.
About the diginomica Network
The diginomica network is an invitation-only forum comprising more than 400 CIOs and CTOs from leading global organizations. It provides a trusted space for technology leaders to confidentially discuss strategic technology challenges.
Free from vendor influence, the network fosters peer-to-peer learning and collaboration among enterprise technology leadership.
ENDS.
