Wipro study finds firms struggle to prove AI value

Wipro and HFS Research have published a study showing that many enterprises are struggling to prove the value of their artificial intelligence spending. The findings point to a gap between AI investment and measurable business results.

According to the research, 87% of enterprises are investing in AI faster than they can demonstrate its value, while 72% lack a consistent way to track enterprise outcomes rather than AI activity alone. It also found that 82% are still running AI in isolated pilots, and only 18% have embedded it across the wider business.

The study adds to concerns in UK boardrooms over whether rising AI budgets are translating into operational gains. It found that 62% of respondents struggle to separate AI activity from real business results, while 65% said spending is driven more by urgency and external pressure than by a clear plan.

Another part of the research examined how businesses are preparing for hybrid teams of people and AI systems. More than half of enterprise leaders surveyed expect such teams to become a reality by 2026, and 90% expect them to become standard within three years.

Yet organisations are not redesigning work quickly enough to support that shift. Nearly a third of enterprises have not designed workflows for shared ownership, and enterprise-wide redesign remains limited, with most activity still confined to pilots.

Workflow redesign

Less than half of respondents said they have AI that is highly contextual or deeply embedded in their operations. The lack of integration makes it harder for companies to show whether AI is improving performance in a meaningful way.

Among organisations running lightly contextual AI, 83% said they struggle to distinguish AI activity from actual business results. That suggests many companies are still measuring deployment rather than impact.

Harsha Anand Almad, Head of People & Change, Consulting, Wipro, said the main barrier is not the technology itself but the lack of broader organisational change.

"AI will not pay off through tools alone-it will only pay off when leaders redesign work, lead the workforce transition, and build operating models fit for human + AI execution," Almad said.

"Yet our survey shows that organisations are falling behind on organisational redesign. For as long as organisations see AI as a 'bolt-on' rather than a 'built-in' capability-contextual and deeply embedded across all systems-proving the real impact of AI will remain elusive. Ultimately, AI is not a tech play; it is about operations, processes and, most importantly, people. Realising its full promise will require an enterprise-wide redesign of people and organisational workflows."

Pressure to spend

The findings suggest executive pressure is shaping investment decisions as much as strategy. Almost two-thirds of respondents said AI spending is being driven by urgency and external pressure, and 69% said they feel pressure to show implementation progress despite uncertain outcomes.

That can leave companies with a collection of pilots and point solutions rather than a coherent operating model. The report argues that this makes scaling harder and weakens confidence in the results reported to senior management and boards.

Amit Kumar, Managing Partner and Global Head of Consulting at Wipro, said companies need to tie AI programmes more closely to the way their businesses actually operate.

"When it comes to driving real value from AI, understanding your processes and building AI to fit your specific context is non-negotiable," Kumar said.

"Organisations need to understand how their business processes run and build AI to remove bottlenecks and solve real, industry-specific problems. True impact is only possible if AI is tailored to a company's unique business context and embedded in an AI-first operating and talent model."

From pilot to production

Phil Fersht, Chief Executive Officer and Chief Analyst at HFS Research, said many companies remain stuck in early-stage AI projects because they have not defined measurable business outcomes from the start.

"Too many enterprises are chasing AI pilots without committing to measurable business outcomes," Fersht said.

"When the race is about what can be shown rather than what can be measured and scaled, organisations rack up fragmented efforts and strategic debt that undermine trust in results. The real inflection point is moving from proof of concept to production. That is when AI stops being a cost lever and starts fuelling growth. But scalable, personalised performance does not happen by accident. It demands a deliberate roadmap to redesign operating models for the AI era."