UK Government Lowers Its Estimate of AI’s Economic Impact
The British government has reduced its estimate of how much artificial intelligence could add to the country’s economy. The newer figure is far below the much larger number that ministers had used in public statements for nearly two years.
Two Very Different Forecasts
The government had promoted the idea that AI could increase the UK economy by £400 billion, or about $538 billion. The figure was used to show that the technology could bring major economic benefits and help ease fears about lost jobs, weaker growth, and a possible economic downturn.
However, the government’s own newer estimate places the likely gain between £55 billion and £140 billion. Its central estimate is £120 billion, which is less than one-third of the earlier claim.
AI minister Kanishka Narayan described the newer figure as a highly uncertain estimate. He also explained that the £400 billion figure came from a public study paid for by a large technology company and prepared by a British consulting firm.
Why the Numbers Changed
The consulting firm’s economics chief, Jonathan Dupont, said the gap between the two forecasts partly reflects the slower spread of AI than was expected when the older study was prepared.
AI is developing quickly, and economic forecasts must change as new information becomes available. The older model assumed that businesses and public agencies would adopt the technology at a faster pace. The government’s newer estimate assumes that adoption will happen more slowly.
The consulting firm still believes AI has strong long-term potential. But it now expects a smaller share of those benefits to appear in the near term. The firm said growth could speed up later if government departments adopt AI more widely and build it into everyday work.
Old Figure Continued to Appear
The £400 billion estimate was used by government departments and public agencies under two different governments. Even after ministers had access to the lower forecast, at least one government agency continued to publish the older number.
The agency is responsible for apprenticeships and job training. Its continued use of the larger figure showed how quickly a forecast can spread through official documents, public speeches, and policy discussions.
The government’s parliamentary answer placed the two estimates side by side for the first recorded time. It also identified the private company that had paid for the original study. However, the answer did not formally withdraw the £400 billion claim or state that it had been false.
Concerns About Hype
Dame Chi Onwurah, a lawmaker who leads a parliamentary committee focused on science and technology, said the government should support useful technology such as AI. She also warned that exaggerated claims could weaken public trust.
Economic forecasts depend on assumptions about business investment, worker skills, public spending, and the speed of technological change. When those assumptions are hidden or presented with too much confidence, the public may receive an unrealistic picture of what AI can deliver.
She said the government should give people a clear explanation of possible savings and benefits instead of relying on figures that may make the technology seem more powerful than it is.
Questions About Research Funding
The debate also focuses on who pays for studies that shape public policy. The original £400 billion estimate was funded by a company that would benefit from wider AI use. The consulting firm said its research remained independent and that its own experts made the judgments and assumptions behind the model.
Still, critics said the government should examine outside research more carefully before using it to support major policies or spending plans.
Sumedha Deshmukh of an independent technology research group said AI forecasts often appear more exact than they really are. She warned that industry-funded studies may contain assumptions that favor faster adoption or larger benefits, even when the researchers act independently.
She urged policymakers to ask three basic questions before using any economic estimate:
- Where did the number come from?
- What assumptions were used to produce it?
- Who is most likely to benefit if the forecast becomes reality?
Calls for Better Evidence
Lord Holmes of Richmond, a Conservative member of the upper chamber, had previously asked ministers to explain the source of the £400 billion claim but did not receive an answer. He said the government should have corrected or explained the figure much sooner.
He also stressed that reliable data matters because public money and government decisions may depend on these forecasts. If the expected economic gains are overstated, taxpayers could support projects that fail to produce the promised results.
The dispute over the £120 billion and £400 billion estimates reflects a larger argument about the proper role of government in promoting AI. Ministers want to encourage investment and show the public that the technology can improve services and productivity. Critics say that encouragement must not become exaggerated promotion.
The central message from the revised estimate is clear: AI may bring major economic benefits, but the size and timing of those benefits remain uncertain. Government officials now face pressure to use independent research, explain their assumptions, and track whether promised gains actually appear.