Insights

AI is continuing to be one of the most significant responsibilities facing data leaders. But as organisations push to adopt it faster, a more difficult question needs to be answered: how can you turn AI ambition into measurable and sustainable business value? 

Our latest report, Chief Information Overload, was filled with comments from data leaders who had found themselves completely responsible for AI by default, with two competing pressures to manage. 

On one hand, there is a huge drive to embrace AI to deliver faster and better outcomes. On the other, strong AI governance and data security are being called for just as loudly. Not forgetting how the accessibility of AI is adding a layer of complication across consistency, control and data quality. 

With expectations and demands continuing to grow, these themes became the focus of discussion at our latest industry roundtable.  

We were joined by David Reed, Chief Knowledge Officer and Evangelist at DataIQ, who opened the session by exploring how grounded leadership can help organisations realise enterprise AI value. His perspective set the stage for a discussion that went well beyond the technology itself. 

Pace of adoption still varies  

Although most organisations have been experimenting with AI for a few years, the conversation highlighted how much variation there is when it comes to moving this into adoption. 

Some organisations are way beyond the experimentation phase, with AI tools and agents fully embedded and delivering value through increased innovation and productivity. Not being able to keep up with the demand is a real challenge for these data leaders.  

Others are taking a cautious approach to AI implementation, favouring controlled rollouts and clear governance frameworks. The time to manage this and oversee quality is the biggest obstacle. 

There are also some who are coming up against clear limitations. With struggles to get buy in from team members who are anxious about their jobs, difficult conversations around continued investment from finance departments who are questioning the impact, and high levels of caution from the C-Suite who are worried about data security.  

AI as an enabler, not a replacement 

Leaders said securing buy-in for AI-enabled ways of working has not been straightforward. As AI agents take on more activities, including some coding tasks, employees are questioning how the skills they have built their careers on will be valued in the future. 

This requires careful change management, but it also creates an opportunity. Automating repetitive or administrative work can free data teams to focus on problem-solving, critical thinking and the decisions where human judgement matters most. 

Leaders described the importance of celebrating early wins, involving teams in the transition and demonstrating tangible improvements in the speed and quality of work. “Adoption is as much a people challenge as a technical one.” 

Data teams of the future  

The conversation turned to the workforce and how AI will influence the makeup of data teams in the future.  

This is something people are already showing caution around. Super users are experimenting and embracing AI as much as possible – too much in some cases, while others are wondering whether they’re training AI to do their jobs and where that could leave them.  

There was unity around the need for dedicated change management support, to help employees adapt, build confidence and ensure responsible adoption. Naturally, with time and any new development, they agreed that there will be changes to what people do or how they do it, which is already happening. 

Questions were also raised about the longevity of specific roles across AI and transformation, and whether they feel temporary as AI moves beyond emerging to another tool in our toolkits. 

AI and the role of the data leader 

There was strong agreement that as AI becomes more embedded, the data leader’s role becomes more important, not less. Governance, accessibility and control remain essential to maintaining confidence in the data behind AI-powered decisions. 

AI is only as good as the data it relies on, but it is also making weaknesses in that data more visible. For data leaders, this creates an opportunity to move the conversation from technical foundations to broader discussions around trust, risk and business value. 

Technical expertise remains fundamental, but data leaders increasingly need the commercial understanding and communication skills to influence stakeholders, solve problems and demonstrate outcomes across the organisation. 

Entry level skills 

One of the most thought-provoking concerns was what happens when AI automates the foundational tasks needed in early careers. If those opportunities disappear, how will the next generation of data leaders develop the judgement needed to challenge and oversee AI outputs? 

The risk of poor-quality or unreliable AI-generated work, leaders described as “AI slop”, reinforces the need for professional standards, critical thinking and meaningful human oversight. 

These were just some of the things keeping data leaders up at night. As entry-level roles change and experienced workers approach retirement, leaders need to consider how they will pass on the expertise that AI cannot reliably replace. 

From experimentation to measurable value 

The roundtable also highlighted a big shift in the commercial conversation. There was shared experience that in the early days of AI, data teams were encouraged to experiment with it as much as possible. Data leaders are now being asked to account for AI spend and demonstrate a return. Without a clear approach, token, infrastructure and subscription costs can quickly build.  

Are super users experimenting more and delivering more or are there areas where AI usage isn’t more optimal than the original way of doing it? These are just some of the questions that teams are balancing.  

The rising costs associated with AI are leading to negotiations with finance teams. From tokens to subscriptions, data leaders are justifying their worth and finding it increasingly difficult to secure continued investment when results and timeframes aren’t guaranteed.  

For teams who are further ahead with experimentation, the focus is now looking at what has been developed and whether it can add value to productivity and operational gains across the wider business. 

While organisations are at different stages of their AI journey, the discussion highlighted a common theme: success depends on more than technology alone. Data quality, governance, leadership, skills and organisational readiness all play a critical role in turning AI investment into measurable business value. For data leaders, the challenge is no longer preparing for AI. It's ensuring their organisation is ready to realise its full potential. 

Our roundtable was hosted by Phil McGarr, Principal Data Solutions Architect.  

The challenges discussed at the roundtable are being faced by data leaders across every sector. Download the Chief Information Overload report here for further insight or get in touch here to find out more about joining a future Data Leaders Roundtable.