Is AI a Bubble? | Gavin Baker on Data Centers, GPUs, and the AI Economy

July 14, 2026

AI Summary

5 min read

Is the current wave of AI investment a bubble? Gavin Baker, managing partner and CIO of Atreides, makes a direct comparison to the year 2000 telecom bust to argue that it is not. His central framework turns on a single metric: dark fiber versus dark GPUs. At the peak of the internet bubble, 97% of laid fiber was unused. Today, by contrast, "there are no dark GPUs." The hardware is running so hot that the biggest problem in a training run is that GPUs are melting. The evidence of real usage and real returns, he argues, makes the bubble narrative a misleading frame.

The Return on Capital Test

The most direct evidence against a bubble, Baker argues, comes from the financials of the companies doing the spending. The biggest purchasers of GPUs are public companies, and since they ramped up capital expenditures on AI, their returns on invested capital (ROICs) have increased by roughly ten percentage points. "Thus far, the ROI on AI has been really positive," he says. This stands in stark contrast to the year 2000, when Cisco peaked at 150–180 times trailing earnings; today, NVIDIA trades at around 40 times. The counterparties writing the checks—the hyperscalers—are collectively generating hundreds of billions in free cash flow annually and hold roughly $500 billion in cash on their balance sheets. As Baker puts it, "I think it's kind of fine" when people worry about a pop.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

Never miss an episode of a16z Show

Get every new episode summarized in your inbox — free, ~5 minutes to read.

No spam. Unsubscribe anytime.

What you'll learn

1 (00:00) **Opening: Is AI a Bubble?** - David George introduces the central question and frames the conversation around infrastructure buildout, economic durability, and long-term value creation.
2 (03:54) **"No Dark GPUs" - Why This Is Not a Bubble** - Baker directly answers the taboo question: we are not in an AI bubble, using a stark comparison to the 2000 telecom bubble.
3 (06:26) **The Counterparty: Why the Buildout Is Financed by the Best Companies in History** - Baker highlights the financial strength of the companies funding the AI infrastructure buildout.
4 (07:46) **The "Round-Tripping" Concern (and Why It's Overblown)** - Baker addresses the fear that AI investment is a circular flow of money, similar to the 2000 bubble.
5 (10:19) **The Model Layer: Market Structure and Humility** - Baker shifts to frontier model companies (OpenAI, Anthropic, Google, xAI), emphasizing the need for humility in predicting winners.
6 (13:10) **The Economics of AI Models: Lower Gross Margins Are Structural** - Baker explains why frontier AI labs will not have the high gross margins of the SaaS era.
7 (14:43) **The Application Layer: Why Low Gross Margins Are a Badge of Honor** - Baker argues that application SaaS companies must embrace lower gross margins to succeed in AI.

+ Full timestamped outline available in the app

Guests on this episode

Gavin Baker

Show Notes

As part of our summer replay series, we're revisiting one of the standout conversations from Runtime, a16z's conference on AI infrastructure and the future of computing.

Gavin Baker, Managing Partner and CIO of Atreides Management, joins David George to examine the biggest questions surrounding today's AI investment cycle. Is AI a bubble? What does the unprecedented buildout of data centers, GPUs, and compute infrastructure mean for the economy? And how should investors think about the companies building the next generation of AI?

The conversation explores frontier models, Nvidia, Google, custom silicon, AI infrastructure, application software, robotics, and why Baker believes today's AI investment cycle looks fundamentally different from the internet bubble of the early 2000s. Along the way, they discuss the economics of GPUs, enterprise software, AI business models, and what comes next as AI moves from experimentation into the broader economy.

Resources:

Follow Gavin Baker on X: https://x.com/GavinSBaker

Follow Atreides Management on X: https://x.com/atreidesmgmt

Follow David George on X: https://x.com/DavidGeorge83

Stay Updated:

Find a16z on YouTube: YouTube

Find a16z on X

Find a16z on LinkedIn

Listen to the a16z Show on Spotify

Listen to the a16z Show on Apple Podcasts

Follow our host: https://twitter.com/eriktorenberg

Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures.

Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.