Anthropic Hunts for AI Computing Power Amid Challenges
Left 67%
Center 33%
Right 0%
2 left · 1 center · 0 right
What happened
Anthropic, the company behind the Claude AI models, has been securing large amounts of computing capacity through deals with technology firms, former cryptocurrency miners, SpaceX, Nscale, Akamai, Volta and others. In an updated S-1 filing that extended results through September 30, it reported quarterly revenue of $17 billion, up from $11.5 billion the prior quarter, while compute costs reached $10 billion and rose to 58 percent of revenue from 48 percent. About $5 billion of that compute was under short-term commitments that cost more than multi-year contracts. The company also described a project called Anthropic Direct aimed at building more of its own infrastructure. Investor Howard Marks separately discussed the difficulty of valuing Anthropic and similar firms given the speed and unpredictability of AI.
Sourcing check · Partly on the record · 3 articles
1 of 3 articles cite a named source for the story. It started with unnamed sources; a named source appeared within the hour.
What these terms mean and how they're counted
- On the record · names a source 1 of 3
- On the record means the article cites a source by name for the story's main claim: a person, or an institution speaking officially, such as a company statement or a named agency's spokesperson. The opposite is an unnamed source. The count is how many of the counted articles cite at least one such source; the names are everyone named across them, each once however the articles spell it. Here: Howard Marks
- Primary evidence · cites a document or data 1 of 3
- Primary evidence means the article cites a record you could check yourself: a court filing or ruling, a dataset or report, a recording or transcript, an official statement or press release, market data. A quote is not a document. The count is how many articles cite one. Here: regulatory filing, chart
- Credits another outlet 0 of 3
- The article says the story comes from another outlet's reporting (“first reported by Reuters”). Those articles relay the story; they are not independent confirmation. Here: Each article presents its own reporting.
- Unnamed sources 1 of 3
- The main claim is attributed to people who are not identified: “people familiar with the matter”, “officials said”. The count is how many articles do this. Here: e.g. “people familiar with the matter”
- How it developed
- When the first report landed and whether it named anyone; when the first named source and the first document appeared. Here: First report Nov 2, 15:08 UTC by New York Times, with no named source. First named source within the hour, from Business Insider. First document the next day, from TechCrunch (regulatory filing).
- Not counted 0
- Opinion, analysis and video pieces, and articles too short to assess. Every count above is out of the counted articles. An article republished under a new headline is one article.
This is about who is cited, not whether the story is true. Each source below says what it names. How we count
ONE-SIDED THIS WEEK.
Only left-leaning outlets have covered this story in the last two days;
Fox Business reported it on 2026-10-26.
Omitted — what each side leaves out
Unpacked
Right-leaning outlets had not covered this as of publication. Left-leaning coverage reports Anthropic hunting scarce computing power through a deal spree with tech giants and former crypto miners. It also carries Howard Marks's warning that the company cannot be valued five years out as it files to go public. Readers on the silent side miss that hunt entirely. The two left accounts barely overlap.
Left-leaning outlets do not even lead with the same story. The hunt coverage opens on the operational scramble, while Business Insider opens on why valuing Anthropic is "so tricky" and never describes the compute scramble.
How much of Anthropic's revenue is being spent on the computing power it is racing to lock up?
Bottom line
A right-leaning reader would miss Anthropic's compute hunt altogether, and a Business Insider reader would get Marks on an unpriceable IPO rather than the Times' deal spree with former crypto miners.
The receipts — 2 facts one side's coverage lacks
- Compute took up 58% of the company's revenue in the new quarter
- The quarter included $5.0 billion of compute under short-term commitments, which currently costs significantly more than multi-year commitments
The Left View
The New York Times framed Anthropic’s activity as a deal spree to obtain one of the AI boom’s scarcest resources. Business Insider focused on Howard Marks’s remarks that projecting Anthropic’s earnings five years out is unusually hard, citing the company’s IPO prospectus language about constantly reinventing its products; Marks called AI inscrutable and unpredictable while noting recent market stumbles had punctured a sense of infallibility, and the piece also referenced caution from Warren Buffett and Michael Burry about AI infrastructure spending. Factual coverage of the updated filing emphasized that compute margins worsened even as revenue grew nearly 50 percent in a quarter, because demand forced Anthropic onto higher-priced short-term neocloud deals, and that compute is expected to remain the largest operating expense.
Our Take (balanced)
The story is substantive: it is anchored in Anthropic’s S-1 and a subsequent filing, named counterparties, specific revenue and compute-cost figures, and on-the-record comments from named investors. The financials, the existence of the deals, and Marks’s statements are established by documents and public remarks; claims about a lasting bubble, future margin improvement, or the durability of demand are interpretive and unproven. No right-leaning outlet has reported this. Watch for the completed IPO documents and roadshow, later filings showing whether compute as a share of revenue falls once more multi-year capacity comes online, and any official responses from Anthropic or its compute providers.
3 sources
More stories
The week's bottom lines, in your inbox
One email a week: the five stories that mattered and what they actually mean. Free.