Companies are reporting returns from AI, but fewer than half say they are scaling that value across multiple business units. A Microsoft-published summary of research commissioned from IDC points to a growing gap between early results and repeatable business-wide gains.
Microsoft AI Watch analysis
What happened
The research surveyed more than 5,200 senior IT and business leaders across 28 industries. Microsoft says 66% of organisations reported substantial or transformative AI results, but only 48% were scaling value across multiple business units. Thirty-six per cent were still piloting AI agents.
The summary says organisations used AI in an average of 6.5 business areas. It also reports that 65% had an established process for governing agents, while 60% were highly confident they could address AI security and governance risks. On workforce preparation, 67% said employees had high trust in AI, but just 20% had a comprehensive organisation-wide skills strategy.
Why it matters
Those figures suggest the challenge is shifting from finding places to try AI to making deployments measurable, governable and useful across an organisation. Microsoft says 71% of organisations expect AI budgets to rise by 10% or more in the next year. If that spending is to produce more than scattered pilots, skills, oversight and a way to measure results need to keep pace.
There is a notable gap between trust and preparedness: reported employee trust is much more common than organisation-wide skills planning. That matters when companies are asking staff to work alongside AI systems, particularly agents that can take actions rather than simply generate text.
Our read
This is a useful snapshot of the awkward middle stage of AI adoption: plenty of activity, less evidence that it is becoming repeatable business value. The figures are based on leader reports, and Microsoft commissioned the study, so treat them as a view of what organisations say, not proof that AI caused the returns. Still, the gaps around measurement, governance and skills are practical questions for any business planning to expand its use.
What to watch
- Whether the full IDC research adds detail on how returns and AI maturity were measured.
- Whether organisations turn agent pilots into wider deployments with clear oversight.
- Whether rising AI budgets come with stronger skills plans and better measurement of outcomes.
Discussion spark: If a company is expanding AI use, should it prioritise workforce skills and governance before adding more tools, or is broad experimentation the best way to learn?
Sources and evidence
- Scaling AI value without losing control: Lessons from Frontier Firms – The Official Microsoft Blog (12 November 2026, 16:09 UTC)
not affiliated with or endorsed by Microsoft