Every AI answer, video stream and online payment runs somewhere physical: in a data center, a building full of servers, storage, network equipment, power systems and cooling. The rapid growth of artificial intelligence has set off a wave of data-center building around the world. For investors, this "build-out" touches many kinds of listed companies, not only technology firms. This article explains who is involved, how they earn money, and what to watch.
What a data center is, in plain words
A data center is a secure building designed to keep computers running all the time. Inside are rows of racks holding servers, connected by fast networks to each other and to the internet. Because servers use a lot of electricity and produce a lot of heat, data centers need a steady power supply, backup generators and powerful cooling.
AI has changed data centers in two ways. First, AI servers use far more electricity per rack than ordinary servers, so new buildings need much more power and stronger cooling. Second, the scale is larger: technology companies are building or renting enormous sites to train and run AI models. Both trends mean more spending on land, buildings, electrical equipment and energy.
Who owns and runs data centers
Large technology companies build many data centers for their own use, to run their cloud services, search, social media and AI products. Their spending plans, announced in quarterly results, are watched closely by investors in the whole chain.
Specialist data-center operators build or buy data centers and rent space, power and network connections to many customers. Some are structured as real estate investment trusts (REITs) in the US, which own property and pay out much of their income to investors. In India, several conglomerates, real estate and telecom companies are building data-center businesses, often with partners.
Cloud customers, from start-ups to banks, rent computing power instead of owning servers, which is one reason demand for data-center capacity keeps growing.
Which kinds of companies benefit from the build-out
A single large data center brings work to many industries. The main groups are:
1. Data-center owners and operators. They earn rent and service fees, usually under long contracts. Their income can be steady, but building new sites needs a lot of money, often borrowed.
2. Power and utilities. Data centers need large amounts of reliable electricity, and AI makes that demand bigger. Power generators, transmission companies and renewable-energy developers can see new demand.
3. Electrical and cooling equipment makers. Transformers, switchgear, backup power, cables and cooling systems are essential inside every site. Companies making this equipment can win large orders.
4. Construction and real estate. Land near power and fibre networks, and the construction work to build specialised buildings, form another link.
5. Servers, chips and networking. The computing equipment itself, including AI chips, servers, storage and network switches, is often the largest part of the spending.
6. Telecom and fibre networks. Data centers must connect to each other and to users through high-capacity networks.
Why location matters. Data centers cluster where power, fibre networks, water for cooling and suitable land come together, and where data rules allow. In India, Mumbai, Chennai, Hyderabad and the Delhi region have become major hubs. Companies with assets or projects in these places often appear in data-center news.
Example on PositoAI: Equinix
Equinix, listed in the US, owns and operates data centers in many cities around the world and rents space, power and network connections to thousands of businesses. It is structured as a real estate investment trust. Its business shows how a data-center owner earns steady, contract-based income, and also how much ongoing investment is needed to keep adding capacity.
On PositoAI (18 Oct 2026): Positivity Score 16,475, ranked #3,140 of 3,999 companies in the US and #5,213 worldwide; market cap about ₹9.2 lakh crore; P/E 63.4; P/B 6.76. These figures change daily.
How data-center businesses make money
Rent and power. Operators usually charge customers for space and for the power their equipment uses. Large customers sign contracts lasting many years.
Interconnection. Customers pay to connect directly to partners, networks and cloud providers inside the same building. These connections can be a valuable, recurring source of income.
Building for one large tenant. Some operators build entire sites for a single technology company, which gives scale but concentrates risk on one customer.
What to watch as an investor
Capital spending and debt. Building data centers is expensive. Watch how much a company spends each year, how it pays for it, and how fast its debt is growing.
Customer concentration. A few very large technology customers account for much of the new demand. If they slow their plans, the whole chain can feel it.
Power availability and costs. In many places, the limit on new data centers is not land or money but access to enough electricity. Rising power costs can squeeze margins.
Supply and pricing. If too many sites are built at once, rents can fall. If supply is tight, operators gain pricing power.
Valuation. Popular themes can push share prices ahead of earnings. Price-to-earnings (P/E) and price-to-book (P/B) ratios, compared with history and peers, show how much growth is already expected. For REITs, analysts also look at cash-based measures of income.
Common misconceptions
"Only tech companies benefit from AI data centers." Power, cables, cooling, construction and real estate companies are also part of the chain.
"Data-center income is risk-free because contracts are long." Contracts help, but debt, power costs, oversupply and customer concentration still matter.
"More AI always means more data centers." More efficient chips and software can reduce the computing needed for each task, which can change how much capacity is built.
Frequently asked questions
Why do data centers use so much electricity?
Thousands of servers run day and night, and cooling them takes additional power. AI servers use even more per rack.
What is a REIT?
A real estate investment trust: a company that owns income-producing property and distributes much of its income to investors.
Is India building data centers?
Yes. India's data-center capacity is growing, led by large business groups, real estate and telecom companies, often with global partners.
Where is data-center demand coming from?
Cloud computing, AI, video streaming, digital payments and government and business digitisation.
How to use this on PositoAI
Search data-center, power, cable and equipment companies on PositoAI to see each one's Positivity Score, rank, ratios, red and yellow flags and an AI summary of recent news. Use Go Deep to check debt and cash flow over time, and Compare to look at two companies side by side. PosiStep 12 of PosiCourse, PositoAI's free stock market course, explains balance sheets and debt.
Suggested reading
What Is a Data Center? · Cooling and Power for AI: Why They Matter · Data Storage for AI · AI Chips Explained · What Are AI Stocks? · What Is a Bond?
More in AI & Markets: How AI Is Changing Indian IT Companies · AI and Electricity Demand · Semiconductor Stocks Explained · How Banks Use AI · Can AI Predict the Stock Market? · Using AI for Stock Research Safely · The AI Bubble Question
Summary
Data centers are the physical home of AI and the internet, and AI is driving a large build-out around the world. Beyond technology companies, the build-out involves data-center operators, power and utility companies, electrical and cooling equipment makers, construction and real estate firms, chip and server makers, and telecom networks. Operators earn rent, power and connection fees under long contracts, but their growth needs heavy investment. Capital spending, debt, customer concentration, power availability, supply and valuation are the key things to watch.