Press Enter to search · ESC to close

Macro

Cathie Wood: US Treasury Yields Are Normal, AI’s 99.99% Cost Collapse Is the Real Story

ARK Invest's Cathie Wood argues that current 10-year Treasury yields are historically normal, with the 1970s-90s as the true anomaly. She highlights a 99.99% annual decline in AI inference costs as a deflationary force markets underestimate, alongside 5.7% money growth not yet reflected in inflation data.

Cathie Wood Challenges the ‘Abnormal’ Rate Narrative

ARK Invest CEO and CIO Cathie Wood used her latest “In The Know” segment to push back on the prevailing market narrative that current US Treasury yields represent a historic anomaly. Wood argued that the 10-year Treasury yield sits near the middle of its long-term historical range, and that the true aberration was not the present moment but the two-decade stretch from the 1970s through the 1990s.

Her framing carries significant weight for crypto and digital-asset investors, who have spent much of the past two years treating elevated risk-free rates as an existential headwind for valuations. If Wood is right that today’s yields are closer to normal than the post-2008 era of financial repression, the sector may need to recalibrate its assumptions about the discount rate environment it will operate in for the coming decade.

AI Deflation: 99.99% Annual Decline in Inference Costs

Responding to Bill Ackman’s concerns about rising interest costs and persistent inflation, Wood pointed to a force she believes markets are underestimating: the collapse in AI inference costs. Inference costs have fallen at an annual rate of roughly 99.99%, a deflationary shock that will ripple through the entire economy’s price structure.

For the crypto industry, this matters in two direct ways. First, decentralized compute and GPU networks — the infrastructure layer that settles machine-learning workloads on-chain — face a brutal margin squeeze as centralized inference becomes nearly free. Second, the same cost curve makes autonomous on-chain agents economically viable at scale, potentially unlocking a new wave of DeFi automation and machine-to-machine payments.

Money Growth and the Missing Inflation Signal

Wood also flagged that money supply growth is running near 5.7% annually, a figure that has not yet surfaced in official inflation data. In her view, this creates a latent risk that could reassert itself if velocity picks up. Crypto assets, particularly bitcoin, have historically been pitched as a hedge against precisely this kind of monetary expansion — a thesis that would gain renewed credibility if inflation surprises to the upside.

Data Center Financing Concentrates in the US

On the infrastructure side, Wood noted that roughly 90% of global data center financing flows to the United States. She also emphasized the signal value of credit default swap charts for hyperscale data center operators, suggesting that CDS spreads may offer an early warning system for stress in the AI buildout’s financing stack.

Forward-looking: If Wood’s thesis holds — normal rates, deflationary AI, and a delayed money-supply effect — the macro backdrop for digital assets may be less hostile than consensus suggests. The key variable to watch is whether AI-driven deflation reaches consumer prices before monetary expansion does. Investors should monitor CDS spreads on data center operators and inference cost benchmarks as leading indicators.

View original

Share
Risk notice This site provides news and information on the crypto, blockchain and Web3 industry for reference only and does not constitute investment advice or any promise of returns. Virtual currency-related activities are illegal financial activities in mainland China; digital asset prices are highly volatile; use at your own risk. This site does not provide trading, token issuance or related referral services.

Related Reading

Latest News

TREE NEWS share card
Long-press image above → Save to Photos / Share
Pitch us Feedback