Gavin Baker - AI Market Jitters - [Invest Like the Best, EP.485]
Gavin Baker discusses the recent market volatility in AI stocks despite improving fundamental metrics, arguing that hyperscalers are under-earning and that operating cash flows will accelerate as GPU contracts reprice higher. He emphasizes that negative sentiment is disconnected from quantitative data showing accelerating demand, token growth, and infrastructure metrics across AI labs and open source models.
Summary
Gavin Baker returns to discuss the July market turmoil that saw AI stocks decline 40-60% despite what he characterizes as materially improving fundamentals. He opens by noting he hasn't encountered a single negative quantitative metric about AI—whether measured by GPU availability, rental pricing, DRAM spot prices, token growth, or infrastructure metrics—yet the market has sold off significantly.
Baker identifies several catalysts for the July decline: Meta's announcement about renting compute capacity (misinterpreted as bearish), the release of Llama 2.1 and Claude, concerns about open source models fragmenting the market, rising real yields and widening credit spreads, and China's reported progress with DUV chip manufacturing. However, he argues most of these concerns are either misunderstood or disconnected from actual demand signals.
A central thesis involves the disconnect between spot pricing and contracted pricing for GPUs. Baker explains that many hyperscalers signed long-term agreements (LTAs) at lower contracted rates when they needed financing. As these contracts roll off and spot prices reset higher, the same installed compute base will generate significantly more revenue despite potential declines in spot prices. This repricing, he argues, will drive dramatic acceleration in operating cash flows, making debt financing unnecessary—solving what would otherwise be a concerning capital cycle problem.
Baker discusses how open source adoption (GLM 4, Llama 3.1, etc.) and routers that distribute queries across multiple models represent margin compression for frontier model providers but don't reduce overall compute demand. A token is a token regardless of which model generates it; all tokens require the same compute, memory, and power. This shifts margins from expensive frontier tokens to cheaper open source tokens, but total infrastructure demand actually increases due to elasticity effects.
The transcript covers the game theory of breaking LTAs, where any hyperscaler who breaks a supply agreement risks devastating allocation cuts if supply dynamics shift back in the vendors' favor. This makes long-term contracts effectively unbreakable despite potential spot price declines, creating durable demand.
Baker addresses regulatory risk as the biggest threat to AI infrastructure growth, noting the industry has done a poor job communicating benefits: data centers actually lower local electricity prices through behind-the-meter deals, create persistent high-wage blue-collar jobs, and have minimal water impact. He advocates for major industry communications efforts to counter misinformation.
The conversation includes discussion of emerging opportunities: continual learning and sample-efficient learning that could reduce training compute demand but remain unproven; NVIDIA's new credit wrapper business model providing revenue sharing; open source inference clouds (Fireworks, Modal, Together) growing rapidly with strong unit economics; and potential for disaggregated inference using SRAM-based accelerators for different model components.
Baker expresses skepticism about the market's interpretation of SpaceX's AI ambitions, noting the company has brought on compute faster and cheaper than established players while maintaining high utilization. He also discusses how the market has failed to price in private company performance (Anthropic, OpenAI, Grok, Cursor) which are accelerating and generating strong returns, leaving NVIDIA trading at historically low forward multiples despite its dominant position.
Throughout, Baker emphasizes the importance of intellectual humility, acknowledging he was wrong about certain things and spent the month stress-testing every assumption. Yet he concludes that underlying fundamentals are materially improving while sentiment remains bearish—creating what he views as an attractive risk-reward setup if hyperscaler cash flow acceleration materializes as expected.
About this episode
My guest today is Gavin Baker, founding partner and CIO of Atreides Management. This is our seventh conversation, and just two months after Gavin's last appearance. It's about the gap between what the market is doing and what companies are seeing. It's been a tough month or so for public AI names, but there's no sign of a slowdown on the ground in Silicon Valley. We discuss the latest moves, contracted vs. spot GPU prices, the game theory of memory supply agreements, and why Claude has become the Walter Cronkite of the stock market. We close on SpaceX, orbital compute, and what Gavin sees as the single biggest risk to all of it. Please enjoy this conversation, from the famous table at Benchmark, with my friend Gavin Baker. For the full show notes, transcript, and links to mentioned content, check out the episode page here. ----- Become a Colossus member to get our quarterly print magazine and private audio experience, including exclusive profiles and early access to select episodes. Subscribe at colossus.com/subscribe. ----- Ramp's mission is to help companies manage their spend in a way that reduces expenses and frees up time for teams to work on more valuable projects. Go to ramp.com/invest to sign up for free and get a $250 welcome bonus. ----- Trusted by thousands of businesses, Vanta continuously monitors your security posture and streamlines audits so you can win enterprise deals and build customer trust without the traditional overhead. Invest Like the Best listeners get a special offer of $1,000 off Vanta when you go to vanta.com/invest. ----- WorkOS is the infrastructure B2B and AI-native companies use to sell to enterprise. It covers everything enterprise security requires: SSO, SCIM, RBAC, Audit Logs, AI governance, and more. Trusted by 2,000+ fast-growing companies, including OpenAI, Anthropic, Cursor, and Vercel. ----- Rogo is the AI platform for finance. They're building agents for Wall Street that are trained to understand how bankers and investors actually do work: from diligence and modeling, to turning analysis into deliverables. To learn more, visit rogo.ai/invest. ----- Ridgeline has built a complete, real-time, modern operating system for investment managers. It handles trading, portfolio management, compliance, customer reporting, and much more through an all-in-one real-time cloud platform. Visit ridgeline.ai. ----- Editing and post-production work for this episode was provided by The Podcast Consultant. Timestamps: (00:00:00) Welcome to Invest Like The Best (00:02:35) First Question: July Was 2022 in a Month (00:04:08) The Private Companies Public Markets Can't See (00:05:06) Old GPUs Repricing Higher (00:06:53) Walking Through the Month (00:08:22) Kimi, GLM 5.2 & the Open Source Freak-Out (00:10:51) Real Yields, Spreads & CDS (00:11:54) Does the Build-Out Need Credit? (00:15:22) A Sell-Off With No Clear Villain (00:17:35) Open Source as Dark Matter (00:18:39) Nvidia's Lowest Forward PE in 10 Years (00:21:35) Claude as Walter Cronkite for the Stock Market (00:23:55) Continual Learning & Sample Efficiency (00:25:19) What Would Actually Scare Him (00:26:38) Routers & the Multi-Model Future (00:30:51) Tokens as a Percent of Comp Spend (00:33:37) The Game Theory of Breaking an LTA (00:36:41) Nvidia's Credit Wrapper & Revenue Share (00:37:45) What He'd Do If He Ran Hynix (00:41:46) Who's More Bullish than Him (00:43:28) China's DUV Machine (00:46:10) Bull Case for Software (00:48:16) The RSI Maximalist View (00:49:31) Inference Clouds Growing Without Burning Cash (00:50:35) The Biggest Risk Is Regulation (00:53:44) Telling the Story Better (00:57:15) Dark Horses (00:58:02) SpaceX in the Public Markets
Key Insights
- Baker found no negative quantitative metrics about AI demand across multiple measures (GPU availability, rental pricing, token growth) despite the 40-60% stock market decline in July, suggesting market sentiment is disconnected from on-the-ground fundamentals.
- Hyperscalers locked into long-term GPU contracts at lower prices while spot prices have risen 50-60% in recent months, meaning the same installed compute base will generate dramatically higher revenue as contracts reprice, without requiring new debt financing.
- Open source models gaining market share shifts margins from frontier model providers (90%+ gross margins) to cheaper open source tokens (30% gross margins), but total compute demand increases rather than decreases because all tokens require equivalent compute resources.
- The game theory of breaking supply contracts makes long-term agreements effectively unbreakable: any hyperscaler that breaks an LTA risks having allocations cut by vendors in future shortage cycles, potentially destroying their competitive position.
- NVIDIA's new credit wrapper and revenue-sharing business model with customers addresses liquidity mismatches in GPU financing while allowing NVIDIA to capture upside through equity stakes and royalties on customer revenues.
- Private AI companies (Anthropic, OpenAI, Grok, Cursor) are accelerating and generating strong returns that aren't visible in public markets, meaning NVIDIA and hyperscaler stocks trade at artificially depressed valuations that don't reflect true underlying demand.
- Inference clouds and router technology enabling companies to distribute queries across frontier and open source models are growing with stronger unit economics than frontier labs, creating a sustainable alternative to expensive proprietary APIs.
- China's reported DUV chip manufacturing capability represents a meaningful but overstated technical achievement—equivalent to going from 2001 to 2026 technology levels—but remains 5+ years away from materially impacting global compute supply.
- The market fixates on open source fragmentation as negative, but the fundamental issue is just a margin shift within the compute consumption stack, not a reduction in total infrastructure demand.
- Regulatory risk through data center moratoria (New York) represents the largest genuine threat to AI infrastructure growth, but the industry has failed to communicate that data centers lower local electricity costs and create persistent high-wage jobs.
- Operating cash flow acceleration from hyperscalers (reported at 28-35% growth adjusting for one-time items) combined with repricing of contracted compute bases will likely fund the entire AI infrastructure buildout without requiring significant debt financing.
- The market's interpretation of specific catalysts (Meta's compute rental, open source progress, DUV machines) has been consistently bearish despite each event either being neutral or positive for actual compute demand when understood correctly.
Topics
Transcript
Ramp is the only platform built to make your finance team leaner, faster, and better, saving businesses 5% annually on average, so you can stay focused on growth. Ramp customers grew revenue 3.2 times faster than the average American business. Visa, Vercel, Cursor, Stripe, Notion, 11Lab, Shopify, and 70,000 other businesses all run on Ramp. Mine does too, and so should yours. Learn more at ramp.com slash invest. Mine does too, and so should yours. Learn more at ramp.com slash invest. Felix by Rogo is a personal finance agent that turns a single prompt into finished client-ready work using your firm's own templates, context, and standards. Send Felix an email like, take these comments and turn them for me, or…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Invest Like the Best with Patrick O'Shaughnessy
Sam Altman - How to Make an Abundant Future - [Invest Like the Best, EP.484]
Sam Altman discusses OpenAI's refocus on core AI development after overextending in 2024, explains the company's early conviction to secure massive compute resources, and outlines his vision for abundant, democratized AI that maintains human agency while addressing critical security concerns like the Hugging Face incident.
Matthew Smith — How America Runs Out of Natural Gas by 2030 - [Invest Like the Best, EP.483]
Matthew Smith argues that the U.S. faces an impending natural gas crisis by 2028-2030 driven by AI data center demand and LNG exports, which will exhaust working gas storage and cause electricity prices to spike dramatically. The problem stems from infrastructure constraints in production, processing, and pipeline capacity rather than resource scarcity, requiring urgent investment in nuclear energy and pipeline infrastructure to prevent economic disruption.
John Kim - How to Raise a Few Billion Dollars - [Invest Like the Best, EP.482]
John Kim, a prolific fundraiser and author of 'The Dow of Fundraising,' explains that successful fundraising is fundamentally about understanding human psychology—specifically the equation 'persuasion equals desire minus fear.' He shares three laws of fundraising (differentiation, trade-offs, and pipeline) and emphasizes that trust, not logic, drives capital allocation decisions.
Jeremy Giffon - The Billion Dollar PDF - [Invest Like the Best, EP.481]
Jeremy Giffen discusses his observations from hundreds of conversations with founders and investors over 18 months, exploring how narrative and storytelling drive capital allocation, the dominance of timeline-native culture in shaping markets and politics, and the emerging 'poster class' as a new priestly class replacing billionaires as arbiters of meaning in society.
Etched - Building AI Hardware to Make Inference Faster and Cheaper - [Invest Like the Best, EP.480]
Etched founders Gavin Uberti and Rob Lockett discuss building the first AI inference chip by a post-ChatGPT startup, their architectural innovations in low-voltage inference and cluster-scale memory, and their philosophy of velocity, vertical integration, and aggressive risk-taking to capture what they believe will become the largest market in the world.