Speaking at Oakland Tech Week on September 28, Federal Reserve Governor Lisa D. Cook mapped three distinct horizons for AI’s economic impact โ and the short-term one points toward more inflation, not less.
In a speech titled “An Update on AI and the Economy” delivered at Oakland Tech Week on September 28, 2026, Federal Reserve Governor Lisa D. Cook set out a three-horizon framework for AI’s macroeconomic role. Her short-term verdict is unambiguous: AI is presently adding inflationary pressure and delaying the Fed’s return to its 2 percent target. Her medium-term concern is a labor market at risk of painful transition.
Her long-term view is genuinely optimistic โ she calls AI “a general-purpose technology, on par with or exceeding breakthroughs, such as the steam engine, electricity, and the internet.”
The anchoring figure is $2 trillion in announced AI investment plans, of which, in Cook’s words, “companies have only spent a small fraction.” She gives no percentage for that fraction, and none is supplied here. The distinction matters: what she describes is not yet spent money. It is announced, which is not the same as committed, and Cook says nothing about how much of it converts into spending or when.
Cook also confirmed that she “voted along with the rest of the FOMC to raise rates 25 basis points at the recent September meeting.” That basis-point figure is hers. She gives the inflation numbers that decision was taken against: total inflation rose an estimated 3.8 percent in the 12 months leading into August, which she calls almost double the target, and core inflation an estimated 3.4 percent.
Unemployment has been trending down over the year, with the most recent reading 4.1 percent in August. Governor Michael Barr addressed related themes in a separate speech in Detroit the following day โ his speech confirmed the FOMC voted unanimously to raise rates but supplied no basis-point number. The two speeches are independent; neither governor mentions the other, and neither is responding to the other.
The key insight: Cook’s transmission mechanism runs through construction labor and energy โ not chips. That is what makes it economy-wide rather than sector-specific, and it is also precisely why she says the Fed’s own tools are the wrong instrument to fight it.
Cook’s own qualifier travels with the top two bars: electricity and water are potentially attributable in part to AI, not attributed to it. The core goods line is the one she contrasts with the pre-pandemic trend, when it was drifting down.
The Structural Read
The most analytically interesting move in Cook’s speech is not the warning โ it is the self-limitation that follows it. She identifies a real price pressure and then explicitly says the Fed should not point its primary tool at it. The passage is worth reading precisely: “attempting to fight sector-specific inflation with monetary policy could be a mistake. Our tools are too blunt to target narrow sectors, and addressing relative price shifts is not our role.”
This creates a structural bind. The inflationary channel she describes is not narrow. Data-center construction competes for the same labor pools and energy capacity as hospitals, housing, and manufacturing. Electricity costs and water costs are each up around 5 percent over the past year โ “potentially attributable in part to AI,” in her words. She does not assert that AI caused those moves, and neither does this piece.
In the same sentence she adds a third series: core goods prices, “which were drifting down before the pandemic, are running over a 3 percent annual pace so far this year.” But she does argue that as AI-specific goods prices moderate, “new and more broadly based price pressures may take its place.” A sectoral boom that graduates into economy-wide input competition is not a sector-specific problem anymore.
There is a second channel: the wealth effect. Cook observes that “a large portion of the rise in equity prices over the past few years can be attributed to enthusiasm about AI, and that added wealth appears to be feeding through to household spending.” That is a demand-side pressure running alongside the supply-side squeeze on inputs. Two inflationary vectors from a single technology cycle โ and the instrument best suited to both (rates) is, by Cook’s own account, too blunt for the more structurally interesting one.
Governor Lisa D. Cook ยท Sep 28, 2026
“Data-center investment relies on inputs, like construction labor and energy, that are broadly used in many sectors in the economy. As a result, increased AI investment could introduce price pressure to those other sectors.”
Three Implications
The $2 Trillion Overhang Is a Price Signal, Not Just a Capital Story
Cook’s framing reorients how to read announced AI capital plans. The economic relevance is not the headline figure โ it is that a large fraction remains unspent, and the inputs Cook names are construction labour and energy. Every project that converts announced plans into active builds adds to the input-cost pressure she describes. On her account it is the pipeline, not the headline total, that reaches the broader economy.
Small Businesses Are the Productivity Canary
The Federal Reserve Small Business Credit Survey she cites found nearly half of small employer firms already using AI, with 71 percent of those reporting increased productivity. Small firms โ more than 99 percent of US businesses, responsible for 61 percent of net new job creation since 1995 โ are the channel through which Cook’s long-term optimism either materialises or stalls. If adoption compresses into large incumbents rather than diffusing down the firm-size distribution, her medium-term labor transition warning becomes more acute, not less.
The Governance Gap Is Now Officially Named
When a sitting Fed governor publicly states that monetary policy is the wrong tool for a price pressure she has just spent a speech documenting, she is also identifying a jurisdiction no one currently owns. The input-market competition she describes โ construction labor, energy โ falls across infrastructure policy, energy regulation, and labor market design, none of which sit at the Fed. Cook’s speech does not propose who should own that problem. That absence is itself the structural signal.
Business Engineer Framework
The Permission Layer
Cook’s speech is a live case study in how the Permission Layer shapes AI’s economic trajectory โ not through direct regulation of the technology, but through the institutional boundaries that determine which levers get pulled and by whom. The Map of AI traces where those boundaries sit across the full stack, from infrastructure to application. Understanding that map is the prerequisite for reading policy signals like this one correctly.
Cook’s speech is not a pessimist’s brief against AI โ her long-term view is genuinely optimistic, and she presents the small-business adoption data to support it. What makes the speech structurally significant is the combination: she maps a specific inflationary transmission mechanism running through construction labor and energy, confirms a 25-basis-point September rate hike, acknowledges that AI-driven equity gains are feeding into household spending, and then states plainly that the Fed’s own instrument is too blunt to address the most interesting part of the problem.
That is an unusually precise public admission of a policy gap, and it is attached to a $2 trillion announced-investment pipeline of which only a small fraction has been spent. Cook’s forecast โ continued inflation pressure in coming months from the AI buildout, higher oil prices, and Middle East supply-chain disruptions โ is hers, not this publication’s. But the structural gap she names belongs to everyone who has to operate inside it.
Every quotation above is verbatim from the text of Governor Lisa D. Cook’s speech, An Update on AI and the Economy, delivered at Oakland Tech Week on 28 September 2026 and published on federalreserve.gov. This publication fetched and read that text directly.The $2 trillion figure refers to announced plans. Cook states that companies have spent only a small fraction of it, she puts no percentage on that fraction, and none is supplied above.
Nothing here should be read as $2 trillion of completed investment.Her own qualifier on the price data is reproduced rather than dropped: electricity and water costs are described as potentially attributable in part to AI. Nothing above asserts that AI raised them, and Cook publishes no estimate of how much of any price series is attributable to the buildout.Cook’s speech of 28 September and Governor Barr’s of 29 September are separate.
Neither mentions the other, neither governor is responding to the other, and nothing above should be read as evidence of a coordinated message. The figure of 25 basis points for the September rate decision comes from Cook’s speech; Barr’s speech states that the FOMC agreed unanimously to raise rates but gives no size. Her expectation of continued inflation pressure in coming months is hers, quoted and attributed, and is not a forecast by this publication.
The finding that nearly half of small employer firms use AI, with 71 percent reporting increased productivity, is from the Federal Reserve Small Business Credit Survey she cites rather than her own estimate. Her medium-term warning that the labour market risks a painful transition is unquantified and undated in the speech. Nothing above predicts anything in this publication’s own voice, and nothing here is investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
