Capital · United States

The AI boom’s missing balance-sheet question: What has actually been verified?

Credited to SiliconANGLE · siliconangle.com · 7h ago

I spent 13 years on the floor of the New York Stock Exchange. One lesson stayed with me long after the trading jackets disappeared: A transaction could be executed, documented and perfectly valid while the assumptions that made it attractive proved less durable than anyone expected.

That distinction matters in the artificial intelligence infrastructure boom. We can believe AI will transform productivity and still ask whether each financing, data-center expansion or long-term capacity agreement is supported by evidence that can survive the investment’s life.

The industry is making commitments across different clocks. Compute contracts may run for years. Debt can mature decades from now.

Servers have much shorter technology cycles. Power infrastructure, transmission upgrades and buildings may outlast several generations of the equipment they were built to support. Meanwhile, enterprise customers are still learning what they will use, how intensively they will use it and what they will pay.

None of this proves the investments are unsound. It does create a question that belongs beside revenue forecasts and capital budgets: What exactly has been verified at the moment an institution decides to rely on a promise?

Contracted demand is not the same as productive utilization

An agreement to purchase compute can be commercially meaningful. It may support financing and demonstrate customer commitment. But it does not by itself establish that the underlying workloads will remain economically productive at the contracted scale, or that the customer will renew on similar terms once technology and prices change.

A serious review must separate the existence of the contract from the economic assumptions behind its duration, usage profile, counterparty capacity and renewal expectations. Contract enforceability and economic durability are related; they are not interchangeable.

Physically deliverable capacity is another milestone

A data-center campus can be announced, financed and partially constructed before all of its necessary electricity, cooling, network and equipment dependencies have become operational. A power purchase agreement is not itself proof of deliverability at the exact commissioning date. GPU orders are not proof that a facility can run the ordered cluster at the planned duty cycle.

For infrastructure planning, I use a simple sequence: requested load, evidenced load, deliverable load, energized load, utilized load and sustained demand. These are not six fancy names for the same megawatt. They describe different claims, each requiring different supporting records.

That distinction is already relevant to public grid decisions. Persistence Analytics Group has submitted comments in MISO’s Large Load Working Group proceedings on load forecasts, application and study processes, paired-resource studies and cost transparency. The recurring question is how planners preserve alignment when a proposed project’s capacity, timing or other material conditions change after an earlier study or commitment.

A guarantee allocates exposure; it does not eliminate it

Financial engineering can provide meaningful protection. Minimum payments, parent guarantees, collateral requirements, termination provisions and residual-value commitments can shift or mitigate losses. They should be evaluated on their actual terms and credit strength.

But a contractual remedy is not the same as the continued usefulness of the underlying asset. A guarantee may support recovery from a counterparty while leaving the investor dependent on that counterparty’s resources, enforceability, timing and the value remaining after a disruption. Those are separate assumptions that deserve separate evidence.

This is especially important when apparently distinct exposures rely on a small set of interconnected customers, infrastructure providers, funding sources or technology assumptions. Diversifying contracts does not necessarily diversify the dependency underneath them.

The missing control is revalidation

Initial diligence usually asks whether a project is supportable at approval. A more durable approach asks what must remain true afterward — and which changes require the original case to be revisited before additional money is committed.

Consider a campus whose customer delays its ramp, whose power availability moves to a later year or whose equipment changes materially. Each event may be manageable on its own. But if the original financing or infrastructure decision assumed the earlier configuration, someone must decide whether the updated evidence still supports the next commitment.

That is a decision gate, not a prediction about whether AI succeeds. It calls for a documented chain: the representation being relied upon; the assumption embedded in it; the evidence supporting that assumption; the dependencies that must hold; the party bearing downside exposure; and the material change that would require reassessment.

Different specialists own different pieces. Engineers assess technical deliverability. Lawyers assess contractual rights. Accountants and finance teams assess financial statements and cash flows. Independent decision assurance asks whether those records, taken together, actually establish the proposition the capital allocator is relying on — and whether known gaps have been made explicit.

The question CFOs should put on the table

The productive-utilization debate is essential. So is the question of who will pay for AI intelligence, and at what price. But before authorizing the next investment, a chief financial officer or investment committee can ask an even more immediate question: Which assumptions are verified facts, which are binding commitments, which are conditional forecasts and which need to be revalidated because something material has changed?

A sector can produce transformational technology and disappointing returns on some of its assets. We have seen that distinction in markets before. The lesson is not to stop investing. It is to stop treating execution, delivery, utilization and durable economic value as if they were the same event.

On the trading floor, the confirmation of a trade was the beginning of understanding the position, not the end. AI infrastructure deserves the same discipline.

Neil P. Osnato is founder of Persistence Analytics Group LLC, an independent Decision Assurance™ and Infrastructure Assumption Verification™ firm. He previously spent 13 years trading on the New York Stock Exchange floor. His work focuses on the documentary evidence supporting consequential infrastructure and financial decisions. He wrote this article for SiliconANGLE.

Selected public background: MISO Large Load Working Group stakeholder feedback (LLWG-2026-7 and LLWG-2026-8), September 2026; The Bond Buyer, “Deciding About Data Centers,” October 28, 2026.

Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.

Are you an AWS customer? Support SiliconANGLE financially by buying your AWS services from our Marketplace portal page and links: https://siliconangle.com/aws-marketplace/

Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.

Atomicwork launches certified AI coworker marketplace for enterprise IT and security

Dropbox acquires Tigris Data for $350M, releases new AI productivity features

Autonomous combat drone maker Shield AI buys radar startup Echodyne for $5.3B

Five9 buys contact center intelligence startup Level AI after posting narrow earnings beat

OpenAI makes new data center commitments, AI policy recommendations

Cloudflare shares surge as revenue growth accelerates to 39% and outlook rises

CLOUD - BY MARIA DEUTSCHER. 14 HOURS AGO

EMERGING TECH - BY MIKE WHEATLEY. 14 HOURS AGO

INFRA - BY MARIA DEUTSCHER. 15 HOURS AGO

Join our community

CapitalUnited States

Why is CoreWeave stock rallying today?

Credited to Investing.com · investing.com7h ago

Useful

View the articleInvesting.com

Useful

View the articleInvesting.com

Front

Capital · United States

Ramp data shows AI spending stalling just as Anthropic nears its IPO

Credited to TechCrunch · techcrunch.com · 6h ago

Companies are using more AI than ever, but they’re not spending more on it.

A new report from Ramp shows AI spending flatlined over late summer even as token volumes rose, a sign that price competition and savvier customers may be capping revenue for OpenAI and Anthropic. According to Ramp’s data, token spend only grew 1.5% across August and September. To compare, in July, it grew 42% from the previous month.

As Anthropic gears up for its IPO, the data suggests investors might be wrong to assume AI companies will keep growing at breakneck speed.

“Almost the entire spend on AI is from venture-backed, high-growth tech companies,” Ramp’s lead economist Ara Kharazian told TechCrunch. “That’s fine as long as AI spend continues to rise, but if it starts to slow down because of something in the market or some other factor, they will all start to slow down simultaneously, and that could lead to other cascading effects.”

Kharazian said he’s already seeing evidence that the top 1% of firms — the most “ AI-pilled ” — have stuttered slightly in their spend in August and September, coming up just below the $10,000 per-employee they were hitting at their summer peaks. A slowdown among that cohort is “sufficient to slow down revenue entirely, despite the fact that we’re seeing rising volumes,” he said.

Ramp data shows growth in spending among median and top 10%, but it’s not enough to make up the difference. Median firms spend $12-$15 per employee per month, and the top 10% spend $600-$1,000.

“If you look at the use cases today, it’s not necessarily clear they will get up to that $10,000 per employee per month mark that the 1% is at,” Khazarian said. While there’s plenty of demand for AI coding agents in industries like software and finance, “it’s not necessarily true that a doctor’s office or a car dealership will need to spend that much on AI.”

“The era of hypergrowth may be over, and now that we know there may not be as much juice to squeeze out of the top 1%,” he continued.

October data did see a re-acceleration, with token spend up 8.9% from September 23 to October 23, but Khazarian said the headwinds on spend growth are strong.

Khazarian (who has published a series of blogs called “ Cracks in the AI Thesis ”) said the slowdown isn’t due to a demand problem. Again, AI volumes are up. Instead he points to the competitive nature of the marketplace.

“OpenAI and Anthropic price cuts have…just made it cheaper to use AI,” he said. In September, OpenAI cut API prices in half for GPT-6 Sol and GPT-6 Luna. Answering the price war, Anthropic has also made significant reductions to pricing for Fable 5.1 and Opus 5.5 — a decision that the company said in a recent filing slashed revenue growth by about $600 million. Both companies have also been offering enterprise discounts that are intended to lock-in spend, but in reality are hitting their revenue targets.

Khazarian said this dynamic means it has very much become a buyer’s market.

Subsidizing use of expensive technology in order to lock in customers has always been a part of the Big Tech playbook. Just recall how cheap it used to be to take an Uber or Lyft ride, and now that there are few other ride-hail players, the cost of going from A to B has risen significantly. Khazarian says those dynamics don’t work in the AI world because there’s just too much competition.

“Meta and Nvidia are also introducing enterprise offerings with pretty cheap models or open source models,” Khazarian said. He also pointed to the range of other open model companies from China and the United States, as well as a new company called Jev — a lightweight tool for choosing between a few predetermined outputs — that is increasingly among the top performers on Ramp.

“For a lot of AI use cases, you just need a simple Yes/No,” Khazarian said. “So you take your already cheap models, you shift them over to Jev, and now your already cheap stuff gets even cheaper.”

The ability to complete workloads at a much cheaper cost is another factor driving the spending slowdown. That’s partly due to the discounts from AI companies, but also because companies are just getting better at being thrifty.

“As you get really advanced in using AI, you get better at figuring out how to control it,” Khazarian said. “You’re more likely to use defaults that prevent usage of very expensive models for long-running tasks. You’re more likely to impose some budgets on employees. You’re more likely to offload some of the spend from standard and light models to even cheaper light models via routers.”

Notion co-founder Akshay Kothari recently posted internal data that matched Ramp’s insights. As he succinctly put it: “tokens go brr. Costs go ahhh.”

tokens go brrr costs go ahhh pic.twitter.com/1whCWCJCyM — Akshay Kothari (@akothari) October 22, 2026

“A lot of the assumptions that we had about the trajectory of AI spend are starting to break down,” Khazarian said. “You can use more AI, but use it for less. That’s great for the customers…but it’s not necessarily great for the model companies. And it does introduce new questions around their revenue trajectories.”

Bloomberg reported Friday that Anthropic’s potential valuation sank from the $2 trillion estimated in early October to just $1.2 to $1.3 trillion. The company is also reportedly expected to receive a high “junk” rating from credit agencies. OpenAI also lowered its proposed valuation to $1.25 trillion from $1.4 trillion and slashed its latest fundraising goals by about $10 billion.

Khazarian says Ramp’s dataset likely informs many investor concerns. There is, however, one potential brightside for spending trajectories, albeit a speculative one.

Much of the thesis for hypergrowth spending is that AI will undoubtedly introduce new use cases that will accelerate things. AI coding emerged just a year and a half ago as the most commercially viable product that has led Anthropic and OpenAI to becoming trillion-dollar companies. The hope among investors is that there will be another revolution that will drive further spend.

Some think personal agents will be the next coding agents. OpenAI, SpaceXAI, and Meta have all released their own personal agents, alongside startups like Instinct. Khazarian is skeptical, though, that personal agents, which are “not as economically productive as coding,” will provide the boost the industry needs for nonstop growth.

“The projection is that there will be new products being introduced that will drive further growth,” Khazarian said. “But based on what we can observe today and the products available and what is commercially advanced, AI spend has started to slow down.”