Warsh maps AI downside risk as investment returns face scrutiny
The Fed chair outlined scenarios, not forecasts. For small B2B sellers, the practical implication is to make customer returns testable rather than promise an AI payoff.
Federal Reserve Chairman Kevin Warsh outlined on November 16 a scenario in which disappointing artificial-intelligence investment returns push equity valuations lower, slow the economy and raise unemployment, according to his prepared remarks.
Speaking at the Productivity and Jobs Task Force Conference in San Francisco, Warsh described contingency planning, not a forecast. Fed staff have begun evaluating possible outcomes using stylized, model-based frameworks, he said.
Growth and employment
The favorable case combines persistently stronger growth with broadly shared gains and an economy close to full employment. A separate scenario pairs faster growth with job displacement: automation expands output, but labor-market frictions leave unemployment above levels consistent with full employment.
In the downside case, expected investment returns weaken after substantial capital spending and stock-market gains. The economy eventually recovers, Warsh said, but without the anticipated productivity gains.
The distinction is between possible paths for the economy, rather than a prediction that AI investment will disappoint.
Financing risks
Federal Reserve Board Governor Michael Barr addressed another potential transmission channel in his financial-stability remarks: AI investment losses could reduce investment and spill over into consumer and business spending.
Barr also discussed debt-financed data centers, complex financing arrangements and off-balance-sheet commitments. Neither speech establishes that customer budgets have already contracted, and Warsh’s scenarios do not establish whether any individual AI offer will succeed.
Sales implications (analysis)
For owners of small B2B firms selling AI services, the inference is conditional: if prospects become more cautious about AI spending or broader investment, a specific return case may earn more consideration than broad promises, with proposals starting from the customer’s problem, identifying the task being improved, establishing its current cost or performance and explaining the intended change.
A small-team seller’s next step is a bounded evaluation with an agreed baseline, relevant costs, measurement period and success criterion, separating measurable savings from uncertain benefits, assumptions from observed results, and released staff capacity from actual cash savings so the expected payoff is understandable and the buyer can judge the result without treating uncertain gains as guaranteed.
Sources
- https://www.federalreserve.gov/newsevents/speech/warsh20261116a.htm
- https://www.investing.com/news/economy-news/feds-warsh-says-emerging-technologies-will-have-significant-impact-but-risks-need-to-be-watched-5030593
- https://apnews.com/article/inflation-warsh-ai-federal-reserve-6f9dc54229bbdd80beb4ee3431a01641
- https://news.bloomberglaw.com/banking-law/feds-warsh-says-staff-are-gaming-out-contingencies-on-ai-impact
- https://www.federalreserve.gov/newsevents/speech/files/barr20261116a.pdf
