Eight in ten UK manufacturers surveyed for Crowe’s latest Manufacturing Outlook Report have adopted AI or are actively trialling it, up from 52% in the firm’s spring 2025 survey. The striking rise suggests AI is moving into more factories, though the figures combine active use with trials and do not show how deeply the technology is embedded.
Watch Desk analysis
What happened
Greater Birmingham Chambers of Commerce reported the survey findings on Friday 9 October. The results also indicate that 72% of surveyed firms are working in defence manufacturing or looking into it. On the practical obstacles, 57% said higher operating costs were their largest brake on growth, while 62% named skills shortages as their toughest recruitment issue.
One manufacturer quoted in the report, Integrated System Technologies chief executive Geoff Archenhold, said his team could now iterate concepts, evaluate designs and generate test code in hours rather than years. That is an example from one company, not a measured result across the survey.
Resultsense’s report on the survey carries the figures and industry comments. The available findings do not state how many manufacturers took part or break down how AI is being used.
Why it matters
The jump from 52% to 80% is a useful signal of growing interest, but “adopted or trialling” covers a wide range, from a small experiment to technology woven into production. It should not be mistaken for evidence that four in five manufacturers are running AI at scale.
The cost and skills figures give that adoption story some necessary context. Firms may be keen to try AI while facing other pressures on investment and recruitment. The gap between testing a tool and making it a dependable part of manufacturing is where the harder questions begin.
Our read
This is a notable survey result, not a victory lap for factory automation. The rise deserves attention, but readers should keep the combined adoption-and-trial measure in view and ask what manufacturers are actually doing with the technology. A percentage can tell us AI is entering the conversation; it cannot, by itself, tell us whether it is improving production.
What to watch
- Whether future surveys separate active AI use from trials.
- What tasks manufacturers are applying AI to, and whether results are measured.
- Whether cost and skills pressures slow the move from experiments to routine use.
Discussion spark: Should manufacturers be judged on how many have tried AI, or only on evidence that it is delivering results in routine production?
Sources and evidence
- 80% of UK manufacturers use or trial AI, Crowe finds – Resultsense (9 October 2026, 08:11 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.