Alphabet's cash flow and Anthropic's filing put a price on AI
This week the bill for AI arrived in public. The companies building it showed investors how much cash the buildout consumes and how much of it now runs on borrowed money, while the companies buying it found that the savings are still mostly a forecast. For any business budgeting AI next year, the question has moved from access to return.
Alphabet's AI spending turns its cash flow negative as Google Cloud doubles
On October 28 Alphabet reported $51.4 billion of capital spending for the third quarter, most of it on data centers and chips. That left free cash flow, the cash remaining after investment, at minus $9.1 billion, its second negative quarter in a row, even as Google Cloud revenue grew 103%.
Investors sold the shares anyway, and on October 30 the Nasdaq closed its worst month since March 2025, Reuters reported. Demand for AI capacity is plainly there; what markets now question is when the spending pays back. Alphabet says it will spend significantly more in 2027, so the pressure on its cash will grow before it eases.
Source: CNBC
Anthropic files for a Nasdaq listing and shows $16.3 billion in half-year revenue
On October 26 Anthropic, the maker of the Claude models, filed publicly to list on Nasdaq under the ticker ANT. First-half 2026 revenue reached $16.3 billion. The headline net loss of $80.7 billion is mostly a non-cash accounting charge on convertible securities sold to early investors; the operating loss was $693 million.
The operating figure is the one to read: an AI developer growing this fast came close to covering its costs, a sign it is not burning cash to grow. The filing also lists the risks Anthropic sees in its business, from chip supply to its largest customers, and any firm signing a multi-year AI contract should read that section.
Source: Yahoo Finance
Wall Street banks and Blackstone line up $63 billion for Anthropic's Broadcom chips
On October 26 three Wall Street banks and the investment firm Blackstone announced $63 billion of financing for AI hardware. A new company will buy Broadcom chips and networking equipment and lease them to Anthropic from 2027, and Broadcom will guarantee part of their resale value.
Anthropic gets its chips without paying up front, and the risk moves to lenders and the chipmaker. The same day SpaceX said it had closed about $40 billion of debt for AI chips, CNBC reported. Computing power is now financed the way airlines finance their fleets, so the cost of borrowing has become part of the price of a model.
Source: Bank of America
Gartner finds only 3% of companies meet their AI cost-savings goals
On October 26 Gartner, the research firm, said in research presented to human resources leaders that 97% of organizations cannot reach their cost-savings targets through AI productivity gains alone. Even so, 85% plan to spend more on AI next year, and 38% are already over their AI budget.
Savings arrive when work is redesigned around the tool, and few firms have done that yet. A board expecting AI to pay for itself in 2027 should ask which roles and processes will change, and by when.
Source: HR Dive
Amazon tells engineers to stop inflating their AI usage
On October 27 Amazon updated its guidance for engineers, calling token counts, the units AI tools are billed by, a "deeply flawed" measure of their work. The guidance says engineers who see no benefit from AI tools should not feel pressured to use them. In May Amazon shut an internal usage leaderboard that staff had gamed.
A business that rewards AI adoption by volume gets volume, and a bigger bill. Amazon now ties AI use to results.
Source: CNBC
OpenAI, Google, Microsoft and Anthropic back one opt-out from AI training
On October 30 the four companies said they will honor one shared entry in robots.txt, the file that tells automated crawlers what they may read on a website. A site that lists the token __usage_ai-training opts out of training by all four. Anthropic applies it now; the other three expect to by 2027.
Once all four apply it, one line covers four of the largest AI developers, which makes the choice cheap for any publisher or brand. It applies only to future training: content already collected stays in the models, and AI search tools that read pages live are governed by separate settings.
Source: PPC Land
Mistral releases Large 4, a model businesses can run on their own servers
On October 29 Mistral, the Paris-based AI company, published the weights of Large 4, the files that make up the trained model, at about one trillion parameters, a rough measure of a model's size. Companies and governments can download it and run it on their own servers, and Mistral is pitching it for coding and cyber defense.
European firms that must keep data inside their own systems now have a large European model they can host themselves. Its performance figures are Mistral's own and compare it only with other open models.
Source: TechCrunch
In the coming weeks Anthropic is expected to set a price range for its shares, and that number will show how much patience investors have left for AI spending. The move for this week is your own: for each AI tool your company pays for, write down the result it should produce by mid-2027 and the person who will check it.
Amazon Demotes Token Counts as Anthropic's Customers Spend 9x More
Anthropic's filing shows business customers spending nine times more; Amazon told engineers token counts are a flawed measure. Buyers should price AI by result.
Anthropic's stock-market prospectus, filed October 26, shows its business customers from a year earlier now spending about nine times as much. A day later CNBC reported that Amazon has told its engineers that token counts, the units AI tools are billed by, are a "deeply flawed" way to measure how they use AI. The seller's best growth figure and the metric a large buyer just demoted measure the same thing: volume. Companies that still judge AI by how much of it they consume are judging it by the seller's number.
The sellers are financed on volume that keeps climbing
Anthropic, the maker of the Claude models, filed to list on Nasdaq with first-half 2026 revenue of $16.3 billion, about fourteen times a year earlier. The figure behind that growth is net revenue retention, which compares what last year's business customers spend now with what they spent then. The prospectus puts it near 900% at September 30, a figure the company says has not been independently verified and that leaves out consumer subscriptions. Existing customers multiplied what they spend.
On October 26 three Wall Street banks and the investment firm Blackstone also announced $63 billion to buy chips from the chipmaker Broadcom and lease them to Anthropic from 2027, with Broadcom guaranteeing part of their resale value, per the banks' release. Those lease payments run for years and are paid from customer spending.
Alphabet showed the other side of that bet on October 28: $51.4 billion of quarterly capital spending left its free cash flow at minus $9.1 billion, CNBC reported, and the company plans to spend significantly more in 2027. This week's Newsletter, Alphabet's cash flow and Anthropic's filing put a price on AI, set out that bill; what it depends on is buyers consuming more every year.
The buyers have started to count something else
Amazon, which is also one of Anthropic's largest investors and sells Claude on its own cloud, now names token usage, AI adoption rates and counts of code changes as poor ways to measure how engineers use AI, warning that chasing them raises costs and complexity, per CNBC. The guidance says an engineer who finds that AI does not help with a task is showing good judgment. Amazon shut an internal usage leaderboard in May after staff gamed it, and a spokesperson said it has not capped token use. Consumption is still allowed; it has stopped being the score.
Only 3% of organizations exceed their cost-savings goals through AI productivity gains, according to research the firm Gartner released on October 26, yet 77% of board members expect AI to deliver savings, and 85% of organizations plan to spend more next year. Gartner analyst CV Viverito said organizations will not generate cost savings from AI at this point.
The clearest outcome-first buyer this week was a government. On October 26 the Dutch cabinet set a target of about 25,000 fewer central-government full-time jobs by 2035, with savings of €900 million a year from 2027, Dutch outlets including NL Times reported. AI is one named lever among several, alongside shared IT systems, and the cabinet has not said how much of the saving it expects from AI. What matters is the order of operations: the result came first, with a number and a date, and any tool can now be measured against it.
The strongest case against
The best objection is that the spending is the value. Nobody forces a company to spend nine times more on Claude; it does so because the tasks get done, and Anthropic's prospectus shows the business close to breaking even on operations in the first half. Gartner's own analyst argues that the larger return comes from moving staff time to new work and growth. On that reading, a missed savings target says little about whether the tool pays.
That is fair, and it marks the limit of this week's evidence. It shows the two measures pulling apart; it does not show that the extra consumption is waste. Amazon's memo covers one company's engineers, and Gartner's figures measure cost savings and say nothing about output quality. A customer spending nine times more may well be getting nine times more useful output. What the evidence does show is that one buyer well placed to judge, an engineering organization the size of Amazon's, is no longer willing to assume it.
We argued in Kroger Cuts Jobs for AI as Bank of America Hires to Fix What It Found that usage grows with every employee who finds the tool useful, and that most business cases leave that cost out. The price is moving in the direction that grows volume. This week OpenAI launched GPT-6.1 Terra, a mid-tier model for agents and coding, at about half the per-token price of its main GPT-6.1 Sol model for developers building on it, the tech-news site BleepingComputer reported. A lower unit price makes more volume easy to approve and harder to question.
Who should be uncomfortable
Finance chiefs and operating heads should be uncomfortable if they report AI progress to their board as licenses activated, tokens consumed or share of staff using the tools. Those are the seller's measures, and Gartner's figures suggest most boards hear them as a promise of savings. Also uncomfortable: companies on usage-metered AI contracts with no result written into them, whose bill rises every time adoption succeeds by the only standard they track.
The move, before the 2027 budget closes, is to write the unit of work next to every AI line: contracts reviewed, tickets closed, invoices matched, hours of outside work no longer bought. Then set a consumption ceiling per unit, and approve growth in spend only where the cost per unit falls. The thing to watch is Anthropic's next filing. Retention near 900% will come down as its customer base matures, whatever buyers do. The signal is the split: if growth from existing customers slows while new customers keep arriving, buyers are trimming volume that produced nothing.
Anthropic has told investors, in its own filing, that the business grows because customers spend more every year, and lenders have lined up $63 billion for chips on the strength of that demand. A buyer has no obligation to share the assumption. Consumption is the seller's revenue line; for the buyer it is a cost line that has to be matched against something it produces. The companies that write that match down now, per tool and per unit of work, will negotiate the 2027 renewals from a position the others will not have: they will know which part of the bill is paying for work.