Half of London firms report an AI-era skills gap

TL;DR:

  • Just 50% of London firms say their workforce has the skills needed for the AI age, down sharply from 63% a year earlier.
  • The share reporting significant skills and capacity gaps hit 15%, the highest the annual BusinessLDN survey has recorded.
  • A record 81% of firms plan to raise training investment, even as a fifth expect to cut headcount.

London’s businesses are racing to adopt AI faster than they can build the skills to use it, according to a Survation poll of 2,043 business leaders for BusinessLDN. The proportion confident their staff had the right capabilities fell 13 points in a year, while the share flagging serious gaps nearly quadrupled to 15%.

Adoption outpaces capability

The data captures a workforce in flux. Three-quarters of firms already use AI in some form and only 5% have no plans to, yet 85% of current users say the technology has changed the skills they need — with demand rising for critical thinking, ethical reasoning and decision-making rather than rote tasks. Some 78% expect a significant need for advanced digital skills within five years, up from 56% in 2023. Sixty per cent of firms with gaps lacked advanced digital skills specifically.

The response has been to invest: a record 81% intend to increase training spend over the coming year. But the picture is not uniformly optimistic. One in five firms plans to cut staff, and among those, a quarter cited reduced demand for entry-level workers because of AI — an early sign of the squeeze on junior roles that recruiters across the UK have begun to flag. The survey ran between late November 2025 and mid-January 2026, against a backdrop of London holding the highest regional unemployment rate in the country.

Looking forward

The findings echo a wider UK pattern in which adoption has hit a tipping point while readiness lags behind. BusinessLDN’s Mark Hilton argued that closing the gap requires “a more agile skills system” responsive to fast-changing needs — a challenge for a training infrastructure built for slower technological cycles. For employers, the harder question is whether reskilling existing staff can keep pace, or whether the gap simply widens as the capabilities firms need keep moving.

Share this article