the diginomica network research reveals change management, not tech, is biggest AI challenge. Can enterprises keep up?

Artificial Intelligence (AI) is a change management issue for organizations, not a technology challenge. Those are the findings of a new piece of research carried out with the Chief Information Officers (CIOs) and Chief Technology Officers (CTOs) who are members of our community - the diginomica network. There are technological obstacles to the adoption of AI in the enterprise, but the most significant barriers involve perceptions, understanding, and collaboration.

AI is being used by 93% of our community and for a wide range of purposes, from the obvious - chatbots - through to drug discovery. The implementation of AI is not the greatest concern of CIOs and CTOs, though. Given the power and possibility AI offers, our community of digital leaders is most concerned with ensuring that the organization follows the right path towards change management. In the past, technology implementations, whether enterprise cloud computing or Software-as-a-Service (SaaS), have only achieved between 10% and 20% of their potential. The reason is that all too often organizations don't follow up the technology implementation with a rigorous change management program of communications and education.

As the diginomica report on CIOs navigating AI's weight of expectation and exploring opportunities states:

This is not a technology implementation, it's a fundamental rethinking of how work gets done, requiring open dialogue about fears, realistic timelines, and sustained commitment to behavior change.

Our community believes AI has the potential to be a seismic change to organizational operating models, but it will only succeed if the change management is central to the adoption:

AI is not a point solution you can simply purchase and deploy. It requires deep organizational change, clean data infrastructure, realistic expectations, proper change management, and senior leadership that can separate genuine opportunity from vaporware.

Failure to adopt and invest in change management will result in AI not delivering any business value. Through change management, organizations will find the ways that AI supports its staff, not replaces them, and in return will realize an increase in productivity, profitability, and customer satisfaction. The good news is that over half of the diginomica network of CIOs and CTOs reports a 50% success rate from their initial AI implementations.

Learning from failure

With 57% of CIOs citing success from AI, there is still considerable room for improvement, and the majority of the community says that AI has failed to meet the "elevated expectations" of their boards and fellow executives. Implementations of Microsoft Copilot, or custom-built AI tools, have failed to deliver a return on investment (RoI) or reduce the headcount of those organizations looking to make layoffs.

One challenge for CIOs and CTOs is that the Large Language Models (LLMs) and AI tools are improving so quickly that proof of concept (POC) projects can fail, but then within months could succeed because the technology has overtaken what was used in the initial POC. This can create frustration between business lines and CIOs. As we reported recently on vendor-led surveys finding signs of poor alignment between CIOs and CEOs, one vendor-led survey is finding signs of poor alignment between CIOs and CEOs, something that the community thought it had put behind it 18 years ago. Chief Executive Officers (CEOs) are looking to AI as a way to combat rising costs in a period of trade war-induced price inflation and to increase productivity. CIOs understand these pressures, and as the report shows, technology alone will not satisfy the CEO's appetite; all members of the senior leadership team (SLT) need to be involved in change management.

This places CIOs and CTOs in a strange position; they need to be skilled at tempering enthusiasm with reality, but also the champions and advocates for where AI can deliver real business value. Digital leaders shared their personal experiences of how they bang the drum for AI, but also have to ask peers to define what they mean by AI. This is done to uncover the requirements a business line leader has, and then, in collaboration, work with them to define if AI is the right technology solution to consider. As stated in the research:

This is not only necessary for technology selection, but also so that CIOs can help business line leaders understand the risks of AI, or any technology, in that role in terms of cybersecurity and data protection. diginomica network members report that agentic AI, generative AI (gen AI), and robotics are commonly confused.

Having to play both the advocate and the voice of reason is a good sign for the Chief Information Officer role, as it shows CIOs and CTOs have become trusted business advisors.

Organizational need

The community shared with us what organizations need to do in order to succeed with AI. The main barrier to AI success is not the technology, but poor data quality. Issues with data quality often stem from legacy systems. The report finds that organizations need real-time data across the business with strong governance and compliance frameworks, as well as a unified data infrastructure:

Without these fundamentals in place, AI investments risk becoming expensive experiments that fail to deliver measurable returns.

Vendors of AI technology will see from the research that there are significant risks to their business model. If AI deployments continually fail to deliver a RoI, investments will dwindle and investors will lose patience in the sector. AI solutions will often be sold to business lines, but this research reveals vendors need a high-quality engagement and relationship with the CIO and CTO community, too.

Our community

The diginomica network is a global community of over 400 CIOs and CTOs who work together to share knowledge and practical insights about enterprise technology. For this research, two debates took place, as well as surveys to discuss and define the challenges and opportunities CIOs and CTOs face with AI. The network serves as a way to understand what's actually happening in the market, cutting through vendor hype to reveal the real-world challenges and opportunities facing technology leaders. You can find all our diginomica network content here.