DePIN is Dead. Long Live Infrastructure-as-a-Service.

DePIN thesis was in building Networks that incentivise ordinary people to crowdsource hardware, bandwidth, and storage, trying to turn resources none of us can negotiate with individually into services someone will actually buy. Part 2 is the evidence of the OG model stopping to work. The pattern, the winners, and where we are on the curve.I. The PatternThe cloud ran the scriptAmazon launched EC2 in 2006, selling raw compute by the hour. Rent a server, configure it yourself. Table 1 shows the infrastructure layer financials for the past twelve years. The elephant in the room is the fact that AWS is 18% of Amazon’s revenue but 57% of its operating income, and Amazon is about to rebuild the infrastructure layer all over again with roughly $200B of CapEx in 2026, driven by AI [2][3].Table 1. AWS by year: the infrastructure layer compounding.YearAWS net revenueAWS operating incomeAWS share of Amazon operating income2013$3.1Bn/an/a2015$7.9B$1.9B~84%2017$17.5B$4.3B>100% (international losses exceeded North America retail profit)2019$35.0B$9.2B63%2021$62.2B$18.5B74%2023$90.8B$24.6B67%2024$107.6B$39.8B58%2025$128.7B$45.6B57%Source: Amazon 10-K segment reporting, 2013-2025 [2][3]The services built on top are the ones getting the value accumulated from the infrastructure, not the infrastructure companies themselves. Netflix pays Amazon an estimated $1B+ a year for effectively all its computing, about 2 to 3 cents of every dollar Netflix earns (an estimate; Netflix delivers video over its own CDN, AWS runs the compute and storage) [4][5]. Snowflake books $4.5B in product revenue and owns zero data centers. Datadog $3.4B, Zoom $4.9B, Airbnb $12.2B, all running on rented infrastructure their customers never see [6][7]. The whole market is renting raw servers by the hour (Infrastructure as a Service or IaaS) against finished software delivered over the internet (Software as a Service or SaaS). Total public cloud spending hit $595.7B in 2024, is forecast at $723.4B for 2025, and passes $1 trillion in 2027 on Gartner’s projection [8].Diagram 1. The service layer and its foundation: worldwide IaaS vs SaaS spending, $B. IaaS is Gartner's vendor-revenue actuals series through 2024; SaaS is end-user spending, with the 2025 forecast dashed. Different Gartner methodologies, so compare shapes, not exact levels [8].The service layer is structurally larger and the absolute gap keeps widening, but the infrastructure layer is currently the faster-growing one. GPUs and power are scarce again, so profits are flowing down to the infrastructure. Likely, if we were able to isolate the GPU demand growth, we would see overall cloud commoditisation even further. Also, if the trend continues, GPUs and power will be an example of the same commoditisation in a couple of years.Telecom runs the same script, just different wiresT-Mobile paid up to $1.35 billion for the parent company of Mint Mobile, a phone brand that owns no towers, no spectrum, no radios, but with Ryan Reynolds [9]. Over 2,100 such carriers operate in about 100 countries, an $89-99B global market of phone companies that rent everything (the industry calls them MVNOs) [10].Table 2.

Tower revenues include international operations, CTIA is US-only; treat the ratio as illustrative. Crown Castle 2024 as reported; $4.46B continuing operations after the 2025 fiber divestiture [12].The same story as in cloud infrastructure, but with a twist. Raw resources get commoditised. The price per user fell 30% since 2010 while data traffic grew 341x. Per unit, infrastructure gets cheaper forever. But the premium sat with contracted, guaranteed capacity. Tower master lease agreements are the original SLA business, and the tower landlords quadrupled revenue on long-term contracts while the carriers fought a price war underneath them. Commoditized layer means price war. Contracted layer means pricing power. That is the lesson DePIN needs.The rule has a nameClayton Christensen formulated this rule down in 2003. When one layer of a stack commoditizes, the attractive profits migrate to an adjacent layer that still offers unique value or solve hard problem [15]. In infrastructure examples so far, that means moving capital to services on top of infrastructure. Different literature has known versions of it for decades, from Perez’s technology cycles to Carr’s big switch [16][17][18][19]; Perez even predicted the sequencing, with the deployment phase arriving after a financial crash, which maps a little too well onto the 2022 and now 2026 crypto bear.Is the same script running inside crypto today?The stage where infrastructure turns invisible is already observable in crypto at large. Kalshi users trade egg prices and NFL games. PYUSD holders see PayPal. The chain underneath is of interest of tech enthusiasts and blockchain purists. Even Hyperliquid is the adjacent case. Its users might know exactly what chain they are on, but they experience an exchange-grade product first and a chain second, and its $844M of 2025 revenue outearned Ethereum itself, though not Solana [20].Table 3. Products where crypto is a feature, not the product.ProductWhat the user experiencesWhat runs underneathKiller metric (2025-2026)KalshiCFTC-regulated prediction market appCrypto deposit rails via Zero Hash (USDC, BTC, SOL)$263.5M revenue in 2025; $22B valuation (May 2026); ~2M MAU [21]HyperliquidCEX-grade perp trading app, chain invisible in the UXIts own purpose-built L1$2.95T volume and $844M revenue in 2025 [20]LayerZero / PYUSDTokens that just work across chains; PYUSD holders see PayPalOmnichain messaging, 90+ chainsPayPal’s PYUSD runs on it; Google Cloud operates a verifier node [22]StablecoinsDollars that settle in secondsPublic blockchains$310B+ supply; $10.2T adjusted 12-month transfer volume (Visa/Allium, Aug 2026 pull); GENIUS Act signed July 2025 [23][24]Every infrastructure sector that matured ran this same script. Crypto is not the exception.

DePIN has no audited reporting standard. As projects move to enterprise deals, revenue becomes a black box. Contracts settle off-chain, disclosures are blog posts, and public narratives can be shaped by whoever writes the listicle. During the audit for this article, one widely circulated figure turned out to be a project’s 2022 funding round recycled as its 2026 revenue.