ع
ITR Economics economist Taylor St. Germain outlined the shift during an October 5, 2026 discussion of the firm’s latest projections. The outlook shows industrial production concluding 2026 with 1.5 percent growth before registering a 0.3 percent contraction the following year. Nondefense capital goods new orders, used as a proxy for business-to-business capital expenditure, are seen ending 2026 at 10.6 percent growth only to slow markedly in 2027. St. Germain stressed that assuming straight-line continuation from current trends represents a dangerous budgeting approach for companies.
Forrester Research predicted in its 2026 technology and security outlook that enterprises will defer 25 percent of their planned artificial intelligence expenditures into 2027 as the gap between vendor promises and delivered value triggers greater financial scrutiny from CFOs. The Cambridge-based firm reported that fewer than one-third of decision-makers can currently link artificial intelligence initiatives to financial growth within their organisations. Chief research officer Sharyn Leaver stated that 2026 marks the end of the artificial intelligence hype period as demands for secure and measurable outcomes intensify.
A PwC global survey of chief executives conducted in 2026 revealed that 56 percent had observed no meaningful revenue or cost benefits from artificial intelligence deployments while just 12 percent reported gains in both areas. Gartner separately determined that only 28 percent of artificial intelligence use cases in infrastructure and operations had fully satisfied return-on-investment expectations with 20 percent failing completely. The research organisation further anticipates that more than 40 percent of agentic artificial intelligence projects may be cancelled by the close of 2027 owing to rising costs, ambiguous business value or insufficient risk controls.
Wells Fargo analyst Ohsung Kwon told clients that 2027 is likely to become a year of artificial intelligence applications rather than infrastructure buildout, with hyperscaler capital expenditure growth set to moderate after peaking. Kwon estimated that artificial intelligence demand must reach approximately 1.4 trillion dollars by 2028 to justify the projected spending levels, equating to about 9,000 dollars per worker annually. The analyst cautioned that additional capital expenditure in 2028 would push the required figure to 2.2 trillion dollars, rendering justification progressively more challenging.
Brookings Institution analysis highlighted potential cash-flow constraints for leading artificial intelligence developers such as OpenAI and Anthropic by 2027, as capital expenditure by major hyperscalers including Microsoft, Amazon, Meta and Alphabet is projected to exceed their combined operating cash flows for the first time in 2026. Pantheon Macroeconomics similarly expects hyperscaler spending growth to ease in 2027 amid supply constraints and higher borrowing costs. These forecasts collectively underscore the necessity for businesses to prepare adaptable models rather than relying on extensions of 2026 frameworks.
The theme that the business built for 2026 will not be enough for 2027 has gained traction among strategists evaluating both macroeconomic shifts and sector-specific technology investments. Companies must therefore evaluate which architectures, vendor relationships and operational assumptions have proven durable over recent years. This recalibration process will determine which initiatives receive continued funding and which are scaled back or eliminated as 2027 unfolds.
