Closing the Gap Between AI Adoption and Impact: Key Learnings from the 2026 Executive AI Forum

AI adoption is spreading rapidly, but measurable financial impact depends on process redesign, accountable ownership and data infrastructure rather than tool deployment alone.

• Nearly nine in ten organizations in McKinsey’s 2026 survey reported regular AI use, while only 37% reported any EBIT impact and just 6% said AI contributed at least 5% to EBIT.
• Summit Partners’ analysis of more than 100 portfolio companies found that those with a named AI owner, a defined constraint and process redesign around it were roughly three times more likely to report a positive, measurable P&L result, although the data were self-reported and the relationship was not causal.
• The article frames the current phase as the middle of a productivity J-curve: after computers were introduced in 1981, U.S. labour-productivity growth rose from roughly 1.5% annually in the early 1990s to 2.6% after 1995, with information technology explaining about two-thirds of the increase.
• A healthcare-services company focused AI on documentation gaps during prior-authorization submissions, reporting 40% fewer requests for additional information and a 28% reduction in turnaround time across more than 12,000 cases.
• OpenAI and Anthropic launched services centred on embedding engineers in customer organisations, while Accenture reported $5.9 billion of generative-AI work in fiscal 2025; Gartner forecast 2026 AI-services spending at $576 billion, above its $461 billion forecast for AI software.
• The eventual productivity payoff remains uncertain: Daron Acemoglu expects less than a 1% total-factor-productivity gain over a decade, whereas Erik Brynjolfsson forecasts much larger and faster gains.

Takeaway: Growth companies should prioritise a measurable business constraint, accountable leadership, process redesign and access to data before expanding AI budgets, and should judge investments by reproducible results net of tool costs rather than adoption rates.