https://www.youtube.com/watch?v=8uoSVG3P6z0
TLDR Anthropic’s IPO filing raises serious questions about whether its AI business can justify more than $800 billion in future compute commitments, especially as growth slows, prices may be falling, and investors would have little control. The discussion also compares autonomous-vehicle leaders, with Waymo far ahead of Tesla in paid rides, and highlights investment opportunities and risks across robotaxis, lidar, fleet services, and autonomous trucking.
When evaluating Anthropic’s S-1, distinguish operating performance from accounting changes. The reported $80 billion first-half net loss was largely linked to an increase in the fair value of convertible notes, rather than cash operating losses alone. However, that accounting explanation does not remove the underlying risks of high spending and uncertain profitability. Investors should review cash flow, operating costs, and financing commitments alongside headline net income.
A company’s latest growth trend can differ sharply from the period emphasized in its IPO materials. Anthropic’s reported sequential revenue growth slowed from 147% in the first half to 45% in the third quarter, while revenue relative to compute and infrastructure costs also declined. The S-1’s headline charts reportedly did not include third-quarter results, making them a backward-looking snapshot. Before assessing an IPO, compare current performance with earlier surges and investigate whether new competitors have changed demand.
Lower prices may attract customers, but they only strengthen a business if the added market share supports sustainable revenue and margins. Steve Eisman said he had heard, but could not confirm, that Anthropic cut prices by about 50% in October to regain share. If a company must rely on steep discounts in a market with few durable competitive moats, large spending commitments become especially risky. Investors should look for evidence that price reductions improve customer retention, usage, and revenue enough to offset lower prices.
Large future infrastructure obligations can weigh on earnings even before the costs appear prominently in financial statements. Anthropic reportedly has more than $800 billion in future compute commitments and expects to increase capital spending, creating uncertainty about the scale of future losses. Assess whether projected revenue can support those commitments and whether the company will need additional financing. For infrastructure businesses such as CoreWeave, also examine borrowing costs: debt near 15% can make a model dependent on repeated borrowing difficult to sustain.
An IPO does not necessarily give public investors meaningful control over a company. Anthropic’s reported share structure gives a single Class F share controlled by its seven co-founders 50.1% of the vote, while a long-term benefit trust helps select the board. Investors should read governance disclosures carefully, including voting rights, board appointment procedures, and the influence of founders or special trusts. Strong founder control may support long-term decision-making, but it can also limit public shareholders’ ability to shape corporate policy.
To compare autonomous-vehicle businesses, look beyond announcements and focus on paid rides, operating locations, safety, and unit economics. Waymo was described as the U.S. leader, operating in 15 cities and providing about 500,000 paid rides each week, while Tesla combines supervised Full Self-Driving with a growing robotaxi effort. Tesla’s planned Cyber Cab production cost of about $25,000 could be an advantage, but scale and real-world performance matter too. Amazon’s Zoox remains earlier-stage, and Uber relies on partnerships rather than its own autonomous vehicles and technology.
Autonomous-vehicle adoption depends on more than the technology itself. U.S. companies must navigate state-by-state approval processes, while driver and labor groups add pressure to the regulatory debate. Public confidence also matters: a Gallup poll found that the share of Americans who believe autonomous vehicles improve road safety fell 12 percentage points from 2018. Investors looking beyond vehicle makers can research lidar suppliers, fleet maintenance and service providers such as car-rental firms, and autonomous trucking companies such as Aurora and Tesla.
Eisman questioned whether Anthropic’s business model is financially viable, pointing to more than $800 billion in future compute commitments and planned capital spending that could weigh heavily on earnings. He also noted that reported revenue growth slowed sharply in the third quarter.
Anthropic’s reported $80 billion first-half net loss was largely an accounting effect related to the rising fair value of convertible notes. Despite reducing losses year over year, its future compute commitments and expected capital expenditures could cause losses to grow substantially.
Reported sequential revenue growth fell from 147% to 45% in the third quarter, while revenue relative to compute and infrastructure costs declined from 2.08 to 1.7. Eisman had heard, but could not confirm, that Anthropic cut prices by about 50% in October. He said such cuts are risky unless lower prices bring enough market share to increase revenue.
He said public shareholders would have little control: a single Class F share controlled by the seven co-founders carries 50.1% of the vote, and a long-term benefit trust helps choose the board. He described this model as “founder worship institutionalized.”
Investors should watch the company’s third-quarter results, which were omitted from the S-1’s headline charts, and assess whether its growth, pricing, and spending commitments support its valuation. Eisman said he had no expectation for the IPO valuation and that the effect of the filing on investor sentiment is uncertain.
CoreWeave borrows to build AI data centers, and its debt costs roughly 15% in interest. Eisman warned that borrowing at that rate may be prohibitive for a business model dependent on continually raising debt.
Waymo leads the U.S. market, operating in 15 cities and providing about 500,000 paid rides per week—roughly seven times Tesla’s weekly mileage. Tesla has expanded quickly and has a cost advantage with Cyber Cabs that cost about $25,000 to produce, but it remains well behind Waymo. Amazon’s Zoox is at an earlier stage, while Uber offers autonomous rides through partnerships.
Tesla’s autonomous business includes supervised Full Self-Driving, used by about 1.7 million customers, as well as its robotaxi service.
The U.S. leads in commercialization, while China has an advantage in manufacturing lower-cost vehicles. The Middle East and Europe have so far leaned toward Chinese companies. In the U.S., fragmented state-by-state regulation means companies must navigate separate approval processes and has fueled lobbying disputes between robotaxi companies and driver and labor groups.
About 40,000 Americans and more than a million people worldwide die in car crashes each year, and 94% of crashes are attributed to human error. However, a Gallup poll found that the share of Americans who believe autonomous vehicles improve road safety fell 12 percentage points from 2018, possibly reflecting broader negative sentiment toward AI.
Suggested areas included U.S. lidar-sensor makers, companies that maintain and service vehicle fleets—including car-rental firms such as Hertz—and autonomous-trucking companies such as Aurora and Tesla.