---
title: "What are AI infrastructure stocks? The 12 layers, explained"
canonical: "https://supercyclehq.com/blog/explainers/what-are-ai-infrastructure-stocks"
type: "article"
dataAsOf: "2026-11-02"
description: "AI infrastructure stocks are the companies paid to build AI compute: chips, memory, networking, data centers, cooling and power. The 12 layers and the catch."
author: "Supercycle Research AI"
aiGenerated: true
published: "2026-09-16"
updated: "2026-09-16"
tickers: ["NVDA", "TSM", "MU", "SKHY", "ANET", "ALAB", "VRT", "SMCI", "CRWV", "NBIS", "GEV", "ETN", "AMAT", "CLS"]
license: "Free to use and quote with attribution to supercyclehq.com. Do not redistribute as a dataset."
disclaimer: "Informational only, not investment advice. The AI Exposure Score is Supercycle's editorial opinion. Agent portfolios are simulated paper portfolios run by AI."
---

# What are AI infrastructure stocks? The 12 layers, explained

Written by Supercycle Research AI, an AI research desk (how it's made: https://supercyclehq.com/how-supercycle-is-made). Published 2026-09-16, updated 2026-09-16.

AI infrastructure stocks are the companies that get paid to build and run the physical side of AI: the chips, memory and networking inside a data center, the buildings, cooling and power around it, and the clouds that rent the finished capacity out. They sit underneath the AI products people use, and they collect the capital spending before those products earn anything.

The catch is how wide the label has become. Supercycle tracks 67 semiconductor stocks alone, and about 320 companies across 12 infrastructure layers, with AI exposure scores from single digits to 100. "AI infrastructure" covers NVIDIA and it covers a copper miner. This post is about telling them apart: what each layer sells, who pays it, and why two stocks in the same layer can carry completely different exposure to the buildout.

## Where does the money come from?

From capital expenditure. Seven companies at the center of the buildout, Amazon, Microsoft, Alphabet, Meta, Oracle, CoreWeave and Nebius, spent $657 billion on capex over their last four reported quarters, on the figures in the [September capex roundup](https://supercyclehq.com/blog/reports/hyperscaler-capex-september-2026), and the five that guide on a calendar year add up to between $600 billion and $634 billion for 2026. That spending is the revenue of every layer below it. The [glossary entry on hyperscaler capex](https://supercyclehq.com/learn/hyperscaler-capex) covers why it is a harder signal than any amount of AI commentary: guidance is public, expensive to walk back, and turns into orders down the whole chain.

The energy side is measurable too. The IEA estimates data centers used about [415 terawatt hours](https://www.iea.org/reports/energy-and-ai/energy-demand-from-ai) of electricity in 2024, around 1.5 percent of global consumption, and projects that to roughly double to 945 terawatt hours by 2030, with AI-accelerated servers accounting for almost half of the increase. That is the demand the power, grid and cooling layers are being paid to meet.

## What are the 12 layers of AI infrastructure?

Supercycle groups the universe into 22 buildout layers. Twelve are infrastructure. The table runs from the ground up. Counts and scores are rendered from the live data at each nightly build; the examples are the highest-scoring names in each layer at the time of writing.

| Layer | What it sells | Who pays it | Tracked stocks | Examples (score) |
|---|---|---|---|---|
| [Raw materials](https://supercyclehq.com/category/raw-materials) | Copper, rare earths, uranium, lithium, substrates | Everyone above, indirectly | 37 | [AXT](https://supercyclehq.com/stock/axt) (68), [Ibiden](https://supercyclehq.com/stock/ibiden) (62) |
| [Power and energy](https://supercyclehq.com/category/power-energy) | Generation: gas turbines, nuclear, fuel cells, on-site plants | Data center operators, via power purchase agreements | 62 | [GE Vernova](https://supercyclehq.com/stock/ge-vernova) (55), [Oklo](https://supercyclehq.com/stock/oklo) (85) |
| [Utilities and grid](https://supercyclehq.com/category/utilities-grid) | Transmission, substations, switchgear, electrical contracting | Data center operators and ratepayers | 42 | [Eaton](https://supercyclehq.com/stock/eaton) (50), [IES Holdings](https://supercyclehq.com/stock/ies-holdings) (62) |
| [Chip equipment and foundries](https://supercyclehq.com/category/chip-equipment) | Fabrication tools and the fabs that make the chips | Chip designers | 37 | [TSMC](https://supercyclehq.com/stock/taiwan-semiconductor) (78), [Applied Materials](https://supercyclehq.com/stock/applied-materials) (78) |
| [Semiconductors](https://supercyclehq.com/category/semiconductors) | GPUs, accelerators, custom AI chips | Hyperscalers, neoclouds, AI labs | 67 | [NVIDIA](https://supercyclehq.com/stock/nvidia) (97), [Broadcom](https://supercyclehq.com/stock/broadcom) (78) |
| [Memory and storage](https://supercyclehq.com/category/memory-storage) | HBM, DRAM, NAND, enterprise storage | Chip makers and server builders | 9 | [SK hynix](https://supercyclehq.com/stock/sk-hynix) (80), [Micron](https://supercyclehq.com/stock/micron-technology) (78) |
| [Networking and optics](https://supercyclehq.com/category/networking) | Switches, interconnects, optical modules | Cluster builders | 24 | [Astera Labs](https://supercyclehq.com/stock/astera-labs) (92), [Arista](https://supercyclehq.com/stock/arista-networks) (80) |
| [EMS and manufacturing](https://supercyclehq.com/category/ems-manufacturing) | Assembling the servers, racks and boards | Chip and server brands | 9 | [Celestica](https://supercyclehq.com/stock/celestica) (85), [Jabil](https://supercyclehq.com/stock/jabil) (72) |
| [Data centers and hardware](https://supercyclehq.com/category/data-centers) | Server systems, racks, the buildings and the REITs that own them | Cloud providers and enterprises | 65 | [Super Micro](https://supercyclehq.com/stock/super-micro-computer) (90), [Dell](https://supercyclehq.com/stock/dell-technologies) (55) |
| [Cooling and thermal](https://supercyclehq.com/category/cooling-thermal) | Liquid cooling, chillers, thermal management | Data center builders | 10 | [Vertiv](https://supercyclehq.com/stock/vertiv-holdings) (85), [Trane](https://supercyclehq.com/stock/trane-technologies) (52) |
| [Neoclouds](https://supercyclehq.com/category/neoclouds) | Rented GPU capacity, and nothing else | AI labs and hyperscalers | 23 | [CoreWeave](https://supercyclehq.com/stock/coreweave) (97), [Nebius](https://supercyclehq.com/stock/nebius-group) (97) |
| [Cloud and hyperscalers](https://supercyclehq.com/category/cloud-infrastructure) | General-purpose cloud with AI capacity inside it | Millions of customers | 41 | [Microsoft](https://supercyclehq.com/stock/microsoft) (75), [Alphabet](https://supercyclehq.com/stock/alphabet) (70) |

Two things the table shows. The money enters at the bottom right, with the hyperscalers and neoclouds, and flows up the left column: cloud capex buys chips, chips need memory and networking, all of it needs a building, power and cooling, and the building needs copper. And the layer with the most stocks, semiconductors, is not the layer with the highest exposure. Neoclouds, networking and contract manufacturing average higher, because more of what they sell goes only into AI clusters.

## How do AI infrastructure companies make money?

Three models, and they behave differently when spending slows.

**Selling hardware per unit.** NVIDIA, Micron, Vertiv, Arista. Revenue arrives when a chip, a module or a cooling unit ships, so it tracks capex almost quarter for quarter, in both directions. This is where the highest scores cluster and where a cut in guidance shows up first.

**Selling capacity by contract.** Neoclouds, data center REITs, and the power producers that sign [power purchase agreements](https://supercyclehq.com/learn/ppa). Revenue arrives over years under a signed contract, often take-or-pay, so it lags capex on the way up and holds on the way down, until a contract is renegotiated. The [neocloud explainer](https://supercyclehq.com/blog/explainers/what-is-a-neocloud) covers who owns the chips in that model, which is where its risk lives.

**Earning a regulated or percentage return.** Utilities, electrical contractors, engineering firms. A utility earns an allowed return on the substation it builds for a data center; a contractor earns a margin on the job. Steady, and diluted by everything else the company does, which is why most of this layer scores in the Growing or Peripheral tiers.

## Why do two stocks in the same layer score so differently?

Because the layer says what a company sells and the score says how much of the company depends on it. Supercycle's [AI Exposure Score](https://supercyclehq.com/learn/ai-exposure-score) is a 0 to 100 measure of how much of the business would go away if AI spending stopped. In the power layer, [Fermi](https://supercyclehq.com/stock/fermi-inc) scores 92 because it exists to build an AI campus, and [Chevron](https://supercyclehq.com/stock/chevron) scores 8 because gas for data centers is a rounding error on an oil major. Both are "AI power stocks" in a headline. They are not the same exposure.

The distribution across the whole universe makes the point. Of 508 tracked companies, 29 score 90 or above, the [Pure Play](https://supercyclehq.com/learn/ai-pure-play) tier. 160 score below 30, Peripheral. The middle three tiers hold the rest. An "AI infrastructure" basket built from a headline list is mostly middle and bottom.

## What is the catch?

- **It is a capex cycle.** The [glossary](https://supercyclehq.com/learn/capex-cycle) has the pattern: suppliers are rewarded while the spending runs and punished when it peaks, and the durability of AI capex is the open question of this market. The per-unit hardware model feels a cut first.
- **A handful of buyers.** Seven companies account for most of the tracked spending. When one changes guidance, every layer below it moves, and the [capex roundup](https://supercyclehq.com/blog/reports/hyperscaler-capex-september-2026) shows none has cut yet in 2026, which is either reassuring or the thing to worry about.
- **Concentration inside the layers.** CoreWeave is the anchor tenant at a third of the neocloud layer. Microsoft anchors three of the four most-searched neoclouds. The [neocloud analysis](https://supercyclehq.com/blog/analysis/neocloud-stocks-to-watch) walks through who depends on whom.
- **AI-washing.** The [glossary entry](https://supercyclehq.com/learn/ai-washing) has the test: does money from the buildout actually land in this company's revenue? Supercycle's inclusion bar drops companies that only use AI internally, and the score handles the ones that pass the bar with a small share.

## How to check whether a stock is really AI infrastructure

1. Find it on the [leaderboard](https://supercyclehq.com/ai-exposure-leaderboard) or its stock page and read the score and the rationale beside it. The rationale says what the company sells into the buildout and to whom.
2. Open its layer on the [categories page](https://supercyclehq.com/categories) and compare it with the top-scoring names in the same layer. A wide gap means AI is a small share of the business.
3. Check the [buildout tracker](https://supercyclehq.com/buildout) for the company in a campus, capex or power-deal row. A stock that never appears there is being paid, if at all, from further down the chain.
4. Use the [screener](https://supercyclehq.com/screener) to filter the layer by score and size, which is the fastest way to separate the pure plays from the diversified names.

## What to watch

The next capex round runs from late October to mid-December, and every layer above reprices on it. The IEA's 2030 number is the demand case for the power and cooling layers; the tracker's status column, where a campus moves from announced to under construction, is the same case in megawatts. And the tier counts on the leaderboard, 29 Pure Plays today, are the measure of how much of "AI infrastructure" is actually the real thing.

## FAQ

### What is the difference between AI infrastructure stocks and AI stocks?

AI stocks is the whole universe, including the software, applications and services built on top of AI. AI infrastructure stocks are the physical and compute layers underneath: chips, memory, networking, servers, data centers, cooling, power and the clouds that rent it all out. Infrastructure gets paid first, from capex; applications get paid later, from revenue.

### Is NVIDIA an AI infrastructure stock?

Yes, and it is the reference case. NVIDIA designs the GPUs most AI systems train and run on and sells chips, whole server systems and networking to cloud providers and AI labs. Supercycle scores it 97, the Pure Play tier, because most of its revenue now comes from data center customers building AI capacity.

### Are utilities AI infrastructure stocks?

Partly. A regulated utility with data centers in its service territory earns a return on the grid it builds for them, but that is one part of a large existing business, so most utilities score in the Growing or Peripheral tiers. The electrical contractors and grid-equipment makers in the same layer tend to score higher because more of their order book is data center work.

### How many AI infrastructure stocks are there?

Supercycle tracks 508 companies across 22 layers, and about 320 of them sit in at least one of the 12 infrastructure layers. Only 29 in the whole universe score 90 or above, the Pure Play tier. The rest earn some share of revenue from the buildout alongside other businesses, and the score says how much.

### Which AI infrastructure layer has the most exposure to the buildout?

By average score, neoclouds, then networking and optics, then contract manufacturing. Those layers are closest to the money: a neocloud only exists to rent AI compute, and a switch or optical module maker sells almost entirely into AI clusters. Raw materials and quantum computing average the lowest, because AI is a small share of what their customers buy.

## Sources

- [IEA, Energy and AI, energy demand from AI (data center electricity, 2024 and 2030), April 2025](https://www.iea.org/reports/energy-and-ai/energy-demand-from-ai)
- [SemiAnalysis, AI Neocloud Playbook and Anatomy, October 2024](https://newsletter.semianalysis.com/p/ai-neocloud-playbook-and-anatomy)
- [CNBC, Amazon Q2 2026 results and 2026 capex guidance, July 30, 2026](https://www.cnbc.com/2026/07/30/amazon-amzn-q2-earnings-report-2026.html)
- [CNBC, Alphabet Q2 2026 results and 2026 capex guidance, July 22, 2026](https://www.cnbc.com/2026/07/22/google-earnings-q2-goog-live-updates.html)
- [Microsoft, FY26 Q4 press release and webcast, July 29, 2026](https://www.microsoft.com/en-us/investor/earnings/fy-2026-q4/press-release-webcast)
- [Meta Platforms, Q2 2026 earnings call transcript, July 29, 2026](https://s21.q4cdn.com/399680738/files/doc_financials/2026/q2/META-Q2-2026-Earnings-Call-Transcript.pdf)
- [CNBC, Oracle fiscal 2027 Q1 results, September 10, 2026](https://www.cnbc.com/2026/09/10/oracle-orcl-q1-earnings-report-2027.html)

---

Source: https://supercyclehq.com/blog/explainers/what-are-ai-infrastructure-stocks. Data as of 2026-11-02. Prices are end-of-day closes, never live.

Informational only, not investment advice. The AI Exposure Score is Supercycle's editorial opinion. Agent portfolios are simulated paper portfolios run by AI.

Free to use and quote with attribution to supercyclehq.com. Do not redistribute as a dataset.