AI Deflation and the Repricing of Growth

Summary

The February 22, 2026 episode of the Jordi Visser Podcast features Jordi arguing that artificial intelligence is structurally repricing growth assets across public and private markets. He contends that AI-driven uncertainty is compressing software valuations, driving extreme dispersion, and shifting capital toward asset-heavy infrastructure and scarce assets such as Bitcoin. Visser links these dynamics to rising credit risk, Chinese open-source AI competition, and mounting labor and household stress in what he sees as a persistent deflationary regime.

Take-Home Messages

1. Structural Dispersion Regime: Extreme stock-level dispersion and a broken covariance matrix signal a new market regime rather than a temporary cycle.
2. Software in an AI Crosshairs: AI agents and bespoke tools are compressing software and SaaS valuations, turning many legacy platforms into potential value traps.
3. Asset-Heavy Beneficiaries: Capital is rotating from asset-light software into asset-heavy sectors tied to AI infrastructure, commodities, and industrial hardware.
4. Credit and Private Market Fragility: Long-duration, illiquid private equity and credit exposures linked to software face mounting stress that could spill over into broader credit markets.
5. Bitcoin and Deflation: In a world of AI-driven deflation and weakening technology benchmarks, Bitcoin is framed as a non-disrupted growth asset that may ultimately lead once software and hyperscalers reprice.

Overview

Jorid Visser opens by asserting that equity indices may still finish the year higher even as the underlying market becomes far more unstable. He points to a surge in stock-level dispersion, with more than twice as many S&P names moving over 15% year-to-date compared with the prior year. In his view, this reflects a structural breakdown of the covariance matrix that will persist for years rather than a short-lived bout of volatility.

He places software and SaaS at the center of this new regime, arguing that AI is eroding their economic foundations. AI agents and bespoke tools can now recreate software functionality in seconds, undermining the long-duration revenue streams that justified premium multiples. Visser characterizes attempts to call a bottom in SaaS as a form of denial, contending that only a small fraction of incumbents will adapt while many become value traps.

To frame the repricing, he draws on Michael Mauboussin’s CAP model, which decomposes valuations into expectations about cash flows, risk, and time. AI, he argues, accelerates industry change and lowers barriers to entry, raising doubts about whether many software firms will still exist even three years from now. These uncertainties bifurcate valuations and realign capital toward asset-heavy beneficiaries of AI infrastructure, including industrials, materials, utilities, energy, and specialized hardware.

Visser extends this reasoning into credit and digital assets, warning that private equity, private credit, and venture capital are heavily exposed to software disruption while lacking liquidity. He highlights emerging stress signals, such as halted redemptions, widening tech-sector spreads, and rising consumer delinquencies, as early markers of a broader repricing of long-duration risk. Against this backdrop, he argues that Bitcoin is not vulnerable to AI disruption and may eventually lead once technology benchmarks underperform, especially as AI deflation squeezes traditional growth equities.

Implications and Future Outlook

Visser’s analysis implies that investors and regulators face a persistent environment of high dispersion, compressed software valuations, and intermittent sharp drawdowns rather than a smooth growth cycle. In such a regime, concentration in legacy software, private technology vehicles, or highly leveraged structures becomes increasingly dangerous as AI accelerates uncertainty and erodes terminal value assumptions. Monitoring sector-specific credit spreads, private market redemption pressures, and the health of long-duration balance sheets emerges as a core stability task.

At the same time, the combination of large hyperscaler capex commitments, intensifying Chinese open-source AI competition, and visible labor and consumer stress sets the stage for more complex policy trade-offs. Cheap, globally accessible AI models can amplify deflationary pressure while shifting innovation geography and undermining pricing power for high-cost providers. In this landscape, the eventual performance of Bitcoin relative to technology indices, and the resilience of household balance sheets, will heavily influence how societies absorb AI-driven deflation and the repricing of growth.

Some Key Information Gaps

1. How should valuation models be adjusted to account for AI-driven uncertainty in long-duration cash flows? Refining these models is essential for pricing risk accurately and avoiding sudden repricings that destabilize both public and private markets.
2. How exposed are private equity, private credit, and venture capital portfolios to AI-induced revenue disruption in software-related holdings? Clarifying this exposure will help assess potential credit contagion channels and inform supervisory priorities for institutions heavily allocated to private markets.
3. Under what macro and market conditions could Bitcoin decouple from technology indices and emerge as a leading growth asset? Understanding these conditions matters for allocators and policymakers evaluating Bitcoin’s role in portfolios during an AI-driven deflationary transition.
4. How much of the global AI workload could migrate to low-cost Chinese open-source models under current regulatory and security constraints? Answering this will shape expectations around AI pricing, standards setting, and the strategic dependence of firms and states on different AI ecosystems.
5. How will AI-induced job displacement and hiring recessions interact with existing household debt burdens? This interaction will strongly influence consumer demand, financial stability risks, and the need for targeted labor and social policy interventions.

Broader Implications for Bitcoin

AI-Driven Deflation and Monetary Policy

AI systems that compress costs, automate knowledge work, and intensify competition can create persistent deflationary forces in prices and wages. Central banks that are calibrated for inflationary shocks may struggle to manage a world where long-duration assets are repeatedly repriced downward and traditional tools have weaker traction. Policy frameworks may need to incorporate explicit assumptions about AI-induced deflation when setting rates, designing liquidity facilities, and stress testing financial institutions.

Repricing Long-Duration Assets and Institutional Portfolios

Pension funds, endowments, and sovereign wealth funds that hold large allocations to private equity, private credit, and growth-oriented public equities are structurally exposed to AI-related uncertainty about future cash flows. If AI shortens business model lifespans and undermines terminal value, existing allocation models and discount-rate assumptions could prove too optimistic. Institutions may need to re-evaluate their reliance on illiquid long-duration strategies, diversify toward assets with clearer scarcity properties, and incorporate explicit AI scenarios into their strategic asset allocation processes.

Bitcoin’s Role in an AI-Deflationary Economy

In a landscape where AI erodes the pricing power of many firms and drives repeated technology repricings, a digitally scarce asset independent of AI production dynamics may acquire new relevance. If traditional technology indices underperform while AI continues to expand, Bitcoin could attract capital as a growth asset that is not directly disrupted by AI agents or model competition. Over a 3–5+ year horizon, this dynamic could reposition Bitcoin from a liquidity-correlated risk asset toward a structural allocation in portfolios concerned with AI-driven deflation and the fragility of corporate equity.

Global Competition in Open-Source AI

Rapid advances in low-cost open-source AI models from multiple jurisdictions can reconfigure where innovation happens and who sets technical and governance standards. Widespread adoption of such models by entrepreneurs and smaller firms may weaken the dominance of a handful of large providers while intensifying the deflation of AI services and software margins. Policymakers will need to balance national security and data protection concerns with the economic advantages of open models, while anticipating how cross-border AI competition interacts with trade, industrial policy, and standards-setting bodies.

Labor Market Transitions Under Agentic AI

Agentic AI systems capable of autonomously executing complex digital workflows threaten to displace not only routine tasks but also a wide range of white-collar roles. If this transition coincides with high household leverage and limited social safety nets, economies may experience prolonged hiring recessions and rising financial stress among workers in disrupted sectors. Over the medium term, education systems, labor regulations, and income-support mechanisms will need to adapt to help workers transition into roles that complement rather than compete with AI capabilities.