Compute is eating the world.
San Francisco, CA
Joined November 2022
- Tweets78
- Following3
- Followers655
- Likes135
Nearly every commodity in history gave rise to a boom fervor and a bust ghost town. The biggest question for compute: How do today's companies, lenders, and investors gain resilience against that volatility?
The open capital market for AI infrastructure will be built by free market competition on capital, development, and operation of intelligence production, not the same five balance sheets as backstops.
Squaretower retweeted
The best way to understand NVIDIA's $500B financing announcement is to read @NecessaryForm's palladium essay.
The core point is that American technological supremacy has never been separable from American financial innovation (the railroads led to the invention of preferred stock and credit ratings; venture capital made the transistor investable; SAFE’s made early stage company formation simpler), but we’re yet to face an asset as difficult as AI compute to underwrite.
To date, AI infrastructure has been underwritten by the highly cash generative MAG7 and venture capital. The instruments and standards that will unlock the next phase do not yet exist.
While this is a step in the right direction for arguably the most important technology of our time, it's still imperfect as NVIDIA is guaranteeing the financing through residual value purchases (for up to 25% of GPU value).
Standardized compute offtake contracts that tranche and distribute the various risks, rating methods, and securitization channels for assets under construction all seem feasible and likely parts of the solution.
The full article by Daniel is in the comments.
His company @squaretower_ are working with AI labs, cloud providers, data center operators, exchanges, and financial institutions to solve this problem.
With the news of Kimi 3, compute procurement is fresh on the minds of many AI infra operators.
Prescience from the piece we released on DeepSeek V4 in May: "The open-weight release calendar can be translated to a compute market calendar. Internally, we have seen sudden shifts in long-tenor capacity costs that could fundamentally alter or even quash the viability of plenty of startups as more companies bid for dwindling supply."
Daniel Painter on how a single open-source release moves the hardware market:
"Literally the week that DeepSeek V4 dropped, the market tightness of H100s went up over 200% in less than a day. And then for the next month or two, the rental rates for H100s increased about 1% to 2% a day."
"In just the week after, the one-year lease rate jumped up over 20% from the data we saw. That was an immediate repricing."
"If you're in procurement at any AI company, or any company that just touches compute, you need to have the open source release calendar as a procurement risk. You need to."
@NecessaryForm @squaretower_
Squaretower retweeted
Capital is abundant in AI, but bankable assets are not.
Read my thoughts on the American AI infrastructure buildout in @palladiummag
AI has the capital intensity of the electrical grid, the depreciation clock of consumer electronics, and the strategic stakes of aerospace. This is not a challenge that America can afford to fail.
Read the full article by @NecessaryForm here: palladiummag.com/2026/07/20/…
Squaretower retweeted
AI has the capital intensity of the electrical grid, the depreciation clock of consumer electronics, and the strategic stakes of aerospace. This is not a challenge that America can afford to fail.
Read the full article by @NecessaryForm here: palladiummag.com/2026/07/20/…
Squaretower retweeted
but commodity pricing also means that tokens are subject to supply and demand.
even the availability of finetuned models are dependent on the availability of the compute to experiment, finetune, and then run inference. in a typical sense, the economic value will flow towards the inference providers. but enterprises that understand this and exert control over the full stack will win.
it all comes back to compute
"This turns compute from a fixed constraint into an actively managed portfolio." - CFO OpenAI, today
"All clouds will need new forms of portfolio optimization that fits their risk profile, capital structure, and financial engineering strategies." - Squaretower Research, three days ago
With @StorkOracle we are able to bring critical datasets to 72+ organizations, enabling Squaretower’s compute data to power new products that underwrite hundreds of billions of dollars in growing compute risk.
Compute is the most critical commodity of the next century. More than ever, we need transparent data and tools to understand how this commodity will shape and form economies of the future. This starts with Squaretower’s proprietary datasets, benchmarks, and research.
If you’re a team interested in data access — whether for understanding compute pricing or for new products — reach out.
Bringing GPU compute pricing on-chain.
@squaretower_ is building the financial infrastructure for the compute economy, aggregating GPU rental pricing across cloud providers and normalizing fragmented configurations into standardized, comparable metrics.
Now, Squaretower joins Stork as a data publisher, delivering high-fidelity GPU pricing directly to on-chain applications, starting with an NVIDIA H100 index.
We've reached a milestone of $20.2 million notional volume traded on our compute index.
AI compute markets are still missing the basics: transparent benchmarks, rapid price discovery, and risk transfer tools, despite GPUs becoming the single biggest and most volatile line item for today's most successful companies. Pricing + hedging exposure to the most important asset of the next century should be straightforward.
Squaretower in @MessariCrypto 2026 thesis for launching the first ever compute derivative — the next wave of startups, AI labs, and infrastructure providers need novel markets & infrastructure to unlock the trillion-dollar opportunity in compute commodities
