The AI Paradox

The fact that AI is now firmly inside the enterprise gets reinforced by the year end compilations from consulting firms and research institutions. The adoption is widespread, pilots are common and confidence in AI’s strategic relevance is indisputable. There is, however, a striking pattern that emerges from these publications. Few findings illustrate the pattern:

- “88% of companies use AI in at least one function, but only 39% see impact on EBIT and most see less than 5%” (McKinsey - Nov 2025)

- "Study of about 1,250 firms finds that 60% are achieving zero material value, 35% scaling but slowly and only 5% achieving value" (BCG - Sep 2025)

- "AI success depends 70% on people and processes, 20% on tech infrastructure and only 10% on algorithms; yet most investment goes to technology” (BCG - Sep 2025)

- "Only 21% fundamentally redesigned workflows" (McKinsey Mar 2025)

- "PwC doesn't just need coders; it needs individuals who can bridge the gap between cutting-edge tech and strategic consulting". Mohamad Ali Kande, PwC.

On paper, AI penetration looks impressive. In practice, value creation remains thin. The questions that increasingly shape AI outcomes can be summarised as under:

- Depth or dispersion - Should AI be applied deeply to selected ‘end-to-end workflows’ that truly move the P&L or started across ‘multiple pilots’ and scaled up based on dividends?

- Top down drive or bottom up discovery - How should organisations balance ‘C suite ownership’ of change with ‘shop floor embrace’ ?

- Speed or capability -  Is it better to move fast through ‘outsourcing & off the shelf tools’, or ‘build capability’ for long term resilience?

- New talent or existing wisdom -  How do organisations combine ‘fresh AI skills’ with the ‘tacit knowledge’ of experienced teams ?

- Big bets or Staging: Should AI be financed as a ‘large upfront commitment’, or ‘expanded in stages’ with value proven before scale?

These questions which concern operating model, leadership, mindset and execution discipline reflect the inherent difficult choices. There is no right or wrong, only limitations of constraints and competition for success. What is undeniable though is that the industries across all sectors are at their inflection point and grappling with the AI dilemma searching for answers.

And then there is a shift in consulting. Studies prove that firms that combine advisory insight with large scale execution are growing faster than those focused primarily on strategy. Translating AI insight into action and embedding it into how work gets done is proving more valuable, than generating the insight itself.

There is no 'one size fits all' solution to the paradox. How organisations confront these questions will determine whether AI becomes another layer of cost and complexity or a durable source of competitive advantage.