The missing AI productivity boom has a name now: the ambition gap. In a report titled The AI Ambition Gap, the University of Manchester’s Productivity Institute examines why heavy adoption of AI tools has not yet produced the forecast productivity gains, and concludes the failure is translational. AI raises productivity on individual tasks; those gains do not automatically become gains across a worker’s whole role or a firm’s output, because the complementary human tasks, judgement, coordination, client interaction and checking, become the binding constraint.
The report’s cited numbers show both sides. Developers using GitHub Copilot completed around 26 percent more tasks in the studies it draws on, and professionals using ChatGPT for routine writing and analysis cut completion time by about 40 percent while improving output quality by 18 percent. Yet a study of AI development tools found coding activity up approximately 240 percent while actual software releases grew only about 30 percent, the gap being the human verification and testing still required before anything ships. Up front costs compound the timing problem: retraining and reorganisation expenses arrive immediately, payback later.
The researchers argue the UK is unusually well placed if the translation is made, with the highest share among six compared countries of workers in occupations combining high AI exposure with high complementarity, reflecting its large financial and professional services sectors. They cite OECD estimates that high exposure economies such as the UK and the US could see annual labour productivity gains of between 0.4 and 1.3 percentage points, and note that even 0.5 to 0.7 points sustained would be a step change against a decade of stalled UK productivity. Lead author Bart van Ark’s summary is that AI’s potential becomes real when faster tasks lead to better performance across the business, which requires investment in management, skills and workflow redesign alongside the technology.
The report, commissioned by IT services firm Bytes Technology Group, arrived the same week as harder edged versions of the same finding: Gartner said 75 percent of corporate board members have yet to see productivity gains or cost reductions from AI, and Apollo Global Management reckons only the biggest tech vendors are seeing a payoff so far. For the sector’s earnings debate, the Institute’s version is the constructive one: the returns are deferred by organisational friction, not absent, but they have to be managed into existence.