Anthropic’s IPO Leak, OpenAI’s Dots vs. Meta’s Muse…

PodcastBig Technology PodcastAlex KantrowitzOct 2

Discussion of Anthropic’s reported revenue expectations, large compute commitments, and customer concentration highlights the tension between rapid AI growth and the risks of fixed spending obligations. Speakers also compare OpenAI’s Dots with Meta’s Muse, weighing enterprise positioning, agent trust, frontier-model costs, and the narrowing gap with open-weight models. In markets, gains remain concentrated in a handful of technology stocks, making the broader AI narrative an important potential driver of index performance and investor appetite for an Anthropic IPO.

Artificial Intelligence Technology Stocks Market Breadth

Preview · 20 more points in the app

Read the Full Summary

Key Takeaways

Anthropic’s projected $100 billion in annualized revenue by year-end is unverified, and the speaker emphasized that the actual figure is unknown; against that uncertainty, the company has $252.4 billion in stated compute commitments over the next decade, 80% of them noncancelable. A speaker identified the mismatch with concentrated revenue and customers that can cut spending as a major risk, while speakers disagreed on whether counterparties would let Anthropic fail.

A speaker said AWS growth accelerated from about 17% before AI to around 30%, attributing much of that growth to dealings with OpenAI and Anthropic. Because the speaker said cloud growth is partly dependent on those companies’ health, the separate view that Anthropic’s trajectory could signal the direction of the broader AI trade has implications beyond Anthropic itself.

The S&P 500 was up around 13% for the year despite weak breadth: Citadel estimated four large technology companies added about 300 index points in Q3, more than the index’s total gain, while other stocks collectively subtracted about 150 points. Speakers warned that this concentration leaves the market vulnerable if belief in the AI story falters; one described a possible 5%–10% decline that could erase the year’s gain, not a firm forecast.

Speakers treated Anthropic’s business performance and potential S-1 disclosure as important tests of the AI narrative, while one argued that a $2 trillion valuation would be difficult to reconcile with $4.5 billion as the latest reflection of its economics. Another questioned whether Anthropic would receive investor enthusiasm comparable to SpaceX’s IPO, given Anthropic’s structural risks.

GEO-POLITICS

California Subpoena — California Attorney General Rob Bonta issued OpenAI an investigative subpoena over cybersecurity risks posed by its AI models.

“Nonstop. Nonstop. Nonstop. That's what Muse sent me this week…”—

TECHNOLOGY

Revenue Expectations — A speaker says a source close to Anthropic—possibly the company or an investor—told The New York Times that the company was expected to reach $100 billion in annualized revenue by year-end. The speaker says revenue of $50 billion–$100 billion would make $500 billion in spending less daunting, while emphasizing that the actual figure is unknown.

“So for instance, anthropic recently, I think it was anthropic, told the New York Times. I mean, we obviously it's like source close to the matter. Right? Was it anthropic?…”—

Compute Commitments — Anthropic’s spending commitments include $111 billion to Alphabet, $110 billion to Amazon, and $31.4 billion to Microsoft over the next decade. Eighty percent of its overall compute commitment is described as noncancelable. Speakers disagree on the risk: one says Anthropic could go bankrupt if it cannot meet its obligations, while another argues counterparties have incentives to prevent its collapse.

“Like like, actually if we outline how this plays out, even again, 80% of that overall compute commit is non cancellable…”—

AWS Growth — A speaker says AWS grew about 17% annually before AI, with its growth rate now around 30%. The speaker attributes much of that growth to dealings with OpenAI and Anthropic, leaving cloud growth partly dependent on those companies’ health.

“But a lot of the growth that we're seeing is coming from their dealings with the open AIs and Anthropics and the health of these companies…”—

MARKETS

AI Market Concentration — The S&P 500 was up around 13% for the year despite its worst breadth since the dot-com bubble, with the median stock below its 52-week high. NVIDIA added more than $1 trillion in market capitalization in 2026, and Goldman Sachs attributed half of that year’s index earnings-per-share growth to AI investment. Speakers warn that the market could be vulnerable if the Anthropic or OpenAI story falters.

“Today's public markets and much of the economy depend on AI's continued growth…”—

IPO Appetite — A speaker says SpaceX’s strong IPO was followed by trading difficulties, although the stock remained above its IPO price. The speaker questions whether Anthropic will draw comparable investor enthusiasm given its structural risks.

“SpaceX set the precedent. Now, of course, SpaceX is a very different company…”—

Index Concentration — Citadel estimates that Microsoft, NVIDIA, Apple, and Meta added about 300 S&P 500 points in Q3—more than 200% of the index’s total gain—while other stocks collectively subtracted about 150 points. A speaker also cites 86 S&P 500 stocks in a bear market and says investors have rotated toward technology and semiconductor names. The speakers argue that a shift in this behavior could be dramatic.

“This is something from Citadel…”—

FORECASTS

S&P Downside — A speaker says that if investors stop believing the AI story, the S&P 500 could give back its 13% gain for the year and fall 5% to 10%. This is presented as a possible scenario, not a firm forecast.

“I'm not gonna say bring the stock market down because that's ridiculous, but could make the stock market sort of revert to where I think it's been trying to point…”—

Anthropic IPO Numbers

Reuters’ leaked figures show explosive revenue growth alongside steep losses, while investors still lack the more current financial disclosures needed to assess the IPO.

Compute Commitment Risks

Anthropic’s large, mostly binding compute contracts create exposure for the company and its cloud partners, though the speakers debate whether counterparties would let it fail.

AI Market Concentration

The speakers argue that index gains and earnings growth are unusually dependent on a small group of AI-linked companies, leaving markets vulnerable to a change in sentiment.

Anthropic IPO Red Flags

Investors must weigh Anthropic’s remarkable growth against fixed spending, customer concentration, losses, and customers’ ability to reduce purchases without long-term contracts.

IPO Stakes For AI

The hosts see Anthropic’s filing as a key test of confidence in the broader AI trade, while noting that a market pullback could follow if growth disappoints.

Meta And OpenAI Competition

They consider whether Meta would undercut OpenAI and Anthropic, concluding that Meta has resources and consumer reach but may struggle to match frontier models in enterprise use.

OpenAI Dots In Practice

The host describes using Dots to draft email, prepare invoices, and plan travel, finding it capable of carrying out multi-step tasks while keeping human approval in the loop.

Agent Trust And Intelligence

The speakers debate what counts as enterprise work, how users will trust always-on agents, and whether frontier intelligence is necessary for routine personal and business tasks.

Open-Weight Model Pressure

They argue that enterprises are spreading usage beyond frontier models as open-weight systems close the capability gap and offer a lower-cost option for many tasks.

Tavis Visual Turing Test

A lifelike real-time video avatar impresses the hosts, who discuss latency, potential customer-service uses, and the heightened risks of scams and deceptive AI identities.