By David Barwick – FRANKFURT (Econostream) – European Central Bank Chief Economist Philip Lane on Monday said artificial intelligence could eventually reshape productivity, investment, employment and the natural rate of interest in the euro area, but that its monetary-policy implications remained too uncertain to support any mechanical conclusion.

In a keynote speech in Frankfurt at the ECB-SAFE-RCEA International Conference on the Climate-Macro-Finance Interface, Lane said ECB staff would continue to deepen their work on AI’s short- and long-run effects, adding that “it is prudent to follow a data-dependent approach in identifying the appropriate monetary policy stance.”

Lane said a standard way to think about AI was as a permanent productivity shock that raised incomes and could push inflation higher early in the transition if households and firms quickly brought expected future gains forward into current spending.

That assumption, however, was unrealistic, he said, arguing that households and firms were more likely to learn gradually about the income and employment effects of AI and adjust spending only slowly, which would make any initial inflationary impulse much weaker.

He also said the implications for the euro area’s natural rate of interest were ambiguous, depending on how strongly AI lifted investment, how income gains were distributed, how much uncertainty households and firms faced, and where AI production and adoption were concentrated geographically.

“[T]he net effect of the AI transition on R* remains uncertain,” Lane said.

According to Lane, the latest staff updates did not show a material change in euro area R* since 2024. He said the upper bound of the ECB’s estimate range had edged up to 2.50% from 2.25%, while the lower bound had stayed at 1.75%.

The speech presented AI as already spreading rapidly through the euro area economy, with ECB survey data showing the share of workers using AI in their jobs rising to 40% in 2025 from 26% in 2024.

At the same time, Lane cautioned that the macroeconomic effects were still modest. He said “there is currently little evidence of a substantial effect of AI on employment in the euro area.”

On productivity, Lane said the eventual gains would depend heavily on the speed and breadth of adoption. ECB staff scenarios, he said, suggested that faster diffusion could produce total factor productivity gains of roughly 0.3-0.4 percentage points per year over the coming decade, while slower adoption would generate smaller gains of around 0.2 percentage points.

He warned, however, that Europe risked falling further behind the United States if adoption remained slower and more uneven, citing weaker digital investment, shallower risk-capital markets and greater dependence on foreign frontier technologies.

The euro area’s bank-centered financial system was not especially well suited to financing AI, he argued, since bank-based funding fit poorly with intangible, long-horizon investment, while venture capital and private-credit channels remained too small.

Still, AI was already changing financial intermediation itself, he said, with most large euro area banks using it in areas such as fraud detection, marketing, chatbots and credit scoring.

The ECB was also deploying AI more deeply in its own work, he said, including inflation forecasting, alternative-data analysis, survey processing and policy preparation, while keeping human review and governance safeguards in place.

Summing up, Lane said the evidence so far pointed to rapid diffusion, rising digital investment and growing market interest, but not yet to a clear macroeconomic break. “The aggregate effects of AI on productivity, employment and inflation remain limited and uncertain at this stage,” he said.