Version 6

Fed Chair Kevin Warsh built his rate policy on AI as a disinflationary productivity force, but that thesis has been described in financial media as having inverted — AI is now doing the opposite of enabling rate cuts.[7] Big Tech has shed $2.7 trillion in market value in June 2026 alone, while AI labs are projected to spend $725 billion on capital expenditure in 2026, a 77% increase from $410 billion in 2025.[8] Accenture, the clearest market proxy for AI disruption of incumbent IT services, is down roughly 50% year-to-date after a ~20% single-day crash, and responded with a $4.18 billion cybersecurity acquisition and a $2 billion buyback boost.[15][16][18][19] Valuation analyst Aswath Damodaran warns that the $10–15 trillion projected AI market requires labor displacement at scale — not productivity augmentation — which directly conflicts with Warsh's deflationary thesis.[22]

The $725B capex projection for 2026 AI infrastructure spending provides a concrete mechanism for how AI is proving inflationary rather than disinflationary in the near term: the investment surge adds demand pressure before any supply-side productivity payoff arrives.[8] If Warsh's rate framework rests on AI-as-productivity-deflator, and AI is instead driving a historically large investment cycle while Big Tech equities reprice sharply, the framework fails on its own terms.

Kevin Warsh, confirmed as Fed Chair in 2026, made AI central to his public economic framework. He declared artificial intelligence 'perhaps the most important economic change' of his lifetime and argued its productivity effects could eventually support lower interest rates.[1][2] A Forbes profile from February 2026 framed his full position as an 'AI, productivity and deflation bet.'[3] To institutionalize this view, he established Fed taskforces covering inflation, data, and AI.[4] By June 2026, financial media had shifted from reporting internal Fed disagreements to framing Warsh more directly as a potential brake on AI-driven markets — one outlet titled its coverage 'Only One Man Can Kill the AI Boom: Fed Chair Kevin Warsh,'[5] while another described how Warsh had 'screwed AI tech beasts.'[6] A Motley Fool piece from May 2026 reported that what Warsh said AI would do — enable rate cuts — it is now doing the exact opposite.[7]

The scale of AI infrastructure investment offers a concrete explanation for that inversion. Major AI labs are projected to spend approximately $725 billion on capital expenditure in 2026, a 77% increase from $410 billion in 2025, and within June alone Big Tech has shed $2.7 trillion in market capitalization, with Goldman Sachs publishing spending trajectory expectations that contributed to investor concern.[8] That capex surge adds near-term demand pressure on energy, hardware, and labor — an inflationary impulse rather than a deflationary one. The US-Taiwan trade relationship reflects the hardware footprint of this investment: US imports from Taiwan have exceeded those from China every month since November 2025, driven by AI semiconductor demand,[9] and a formal US-Taiwan trade and investment agreement signed in early 2026 featured AI and supply chains prominently.[10][11] SemiAnalysis data from May 2026 shows US headline industrial production rose only 0.1% with manufacturing flat, while semiconductor production continued upward — AI infrastructure is the only positive driver in an otherwise flat manufacturing sector.[12] SemiAnalysis separately cautioned that the import reversal headline 'means both more and less than it appears,' given how AI investment has distorted traditional trade accounting metrics.[9]

The Accenture selloff is the sharpest market signal that investors are repricing AI's disruption to incumbent IT services. The stock fell approximately 20% on June 18, reaching its lowest price since 2017,[13][14] with the year-to-date decline reaching approximately 50% by late June.[15] WSJ and Yahoo Finance covered both the earnings deterioration and investor uncertainty about AI's effect on Accenture's core consulting model.[16][17] Accenture responded with a $4.18 billion cybersecurity acquisition[18] and a $2 billion boost to its buyback program.[19] A contrarian piece argued AI is accelerating consulting demand and the disruption thesis is overstated,[20] but Accenture's capital allocation — returning cash rather than investing organically — and its continued stock decline cut against that reading.

Valuation analyst Aswath Damodaran has offered two related structural observations. AI companies have real revenues and business models unlike dot-com era firms, a distinction that resists simple crash predictions.[21] But Damodaran has also called the $10–15 trillion projected AI total addressable market 'terrifying' rather than bullish: achieving a TAM of that scale requires AI to replace human labor at scale, not merely enhance productivity.[22] This distinction connects directly to Warsh's thesis — if AI success requires labor displacement rather than productivity augmentation, the deflationary story underpinning his rate policy does not hold. The $725B capex projection adds a second layer to the same problem: even if AI eventually delivers productivity gains, the investment phase itself runs hot.