Where AI Lives: Inside the Hyperscale Data Centers Powering the AI Boom is a book by Scott Stillings, published by Data Center Strategies as part of the Inside Hyperscale series. It explains, in plain language, how hyperscale data centers work, who operates them, and how they power AI, from power and cooling to the people on shift.
Facts about the book
| Field | Value |
|---|---|
| Title | Where AI Lives: Inside the Hyperscale Data Centers Powering the AI Boom |
| Author | Scott Stillings |
| Series | Inside Hyperscale |
| Publisher | Data Center Strategies |
| ISBN | To be announced |
| Publication date | To be announced |
| Formats | Ebook, paperback, hardcover, and audiobook (planned) |
| Page count | About 250 to 270 pages (ESTIMATE, expected length at 6 by 9 trim from the November 8, 2026 draft of 58,675 words; the final count depends on the finished manuscript). |
| Price | To be announced |
| Language | English |
| Retailers | To be announced |
What is a hyperscale data center?
A hyperscale data center is a very large data center campus, one or more buildings plus shared support infrastructure such as substations, generators, a cooling plant, and operations buildings, built to run computing at a scale where a single campus can draw tens or hundreds of megawatts. The hierarchy inside it runs campus → building → data hall → row → rack → server. A data center is not a warehouse, though it may have a warehouse-like appearance; it is a working industrial facility whose product is computation, and whose by-product is heat that must be removed continuously.
The scale is real and measured. Synergy Research Group counted 1,360 hyperscale data centers at the end of Q4 2025, with hyperscale operators accounting for 48% of worldwide capacity (VERIFIED; Synergy Research Group, 7 April 2026). CBRE counted 7,481 MW under construction in H1 2026 across its eight primary North American markets, up 24.8% year over year, with 80.4% preleased (VERIFIED; CBRE, North American Data Center Trends Report, H1 2026, published August 2026). The International Energy Agency's base-case forecast puts global data-center electricity at about 950 TWh by 2030 (FORECAST; IEA, Key Questions on Energy and AI, 16 April 2026).
What does "where AI lives" mean?
AI does not live in the cloud as an abstraction. It lives in physical buildings, on racks of servers, drawing power and producing heat, watched over by operators on shift around the clock. This book is about those buildings, those people, and the systems that keep them running, power in, heat out, nothing down.
Who is this book for?
Operators and managers who run critical facilities; engineers and students who want to understand how hyperscale data centers actually work; business readers who want to know what sits behind the AI boom; and curious readers with no technical background. It is written in plain language first. It is not a technical manual, and it is not a vendor guide.
What the book covers
The manuscript is a Prologue, 14 chapters, and an Epilogue: about 58,700 words in the draft of November 8, 2026; the count will be updated on publication. Chapter titles below are the manuscript's contents and may change before publication.
- Prologue: "3 A.M."
- 1, The Cloud Has a Basement
- 2, Where the Internet Actually Lives
- 3, Follow the Electron
- 4, Power That Cannot Blink
- 5, The Heat Problem
- 6, When Air Isn't Enough
- 7, The Nervous System
- 8, Built to Fail Safely
- 9, The Humans in the Loop
- 10, Procedures Are the Product
- 11, When Things Go Wrong
- 12, Why Here?
- 13, The Real Costs
- 14, Where AI Lives Next
- Epilogue: The Next Question You Ask
The book draws on the author's 20+ years in critical infrastructure operations, U.S. Navy submarine reactor operator, commercial nuclear operator and senior-reactor-operator-certified instructor, and hyperscale data center critical operations manager, and on the shared base understanding that underpins this site: electrical, cooling, controls, fiber, fire protection, and physical infrastructure, described as common patterns and teaching archetypes, not as a single universal design. There is no one unifying design for a data center, and a deviation from a common pattern is not automatically a defect. A chapter-by-chapter walk through the book is Inside Hyperscale: A Reader’s Guide to Where AI Lives.
Citable facts from the book's subject
Every figure the book uses about the industry is labeled the same way it is on this site. A few of the facts the book explains:
- Power is the leading cause of impactful outages for the fifth consecutive year in the 2026 analysis (VERIFIED; Uptime Institute, Annual Outage Analysis 2026, May 2026).
- 47% of operators reported an impactful outage in the prior three years, down from 50% in 2025, the sixth consecutive year of decline on a per-site basis (VERIFIED; Uptime Institute, Global Data Center Survey 2026, July 2026).
- The industry fleet average PUE is 1.52 (VERIFIED; Uptime Institute, Global Data Center Survey 2026, July 2026). PUE is total facility energy divided by IT energy, and must be stated with its measurement boundary, period, IT load, and climate.
- Average rack density is rising: AFCOM's 2026 survey puts average rack density at 27 kW, up from 16 kW the year before (VERIFIED; AFCOM, State of the Data Center 2026, early 2026).
- Air cooling struggles above roughly 30–40 kW per rack; the ceiling is design-dependent, not a universal rule (VERIFIED, range across sources).
The book's own claims are not presented here as external facts. Any claim about hyperscale scale, power, PUE, or AI demand on this page cites a primary source with a link and a date.
Who is Scott Stillings?
Scott Stillings, Ph.D. Business, M.B.A., B.S. Nuclear Engineering Technology, has spent 20+ years in critical infrastructure operations: U.S. Navy submarine reactor operator and senior nuclear maintenance technician (2002–2008), commercial nuclear operator, reactor operator, senior-reactor-operator-certified instructor, and nuclear operations training manager, and now a critical operations manager for a large hyperscale data center campus. He is the author of Where AI Lives: Inside the Hyperscale Data Centers Powering the AI Boom and the founder of Data Center Strategies.
More about Scott → · About Data Center Strategies →
Where to buy
Publication details and retailers will be listed here when confirmed. Until then:
To ask about availability, email info@datacenterstrategies.com.
Scope
Scope: this page describes the book and its contents.
Frequently asked questions
What is Where AI Lives about?
It explains, in plain language, how hyperscale data centers work, who operates them, and how they power AI, from power and cooling to the people on shift.
Who wrote it?
Scott Stillings, a critical-infrastructure operator with 20+ years of experience in the U.S. Navy, commercial nuclear power, and hyperscale data centers. He holds a Ph.D. in Business, an M.B.A., and a B.S. in Nuclear Engineering Technology.
Who is it for?
Operators, managers, engineers, students, business readers, and curious readers. It is written for people who want to understand the buildings behind AI, not only for engineers.
Is it a technical manual?
No. It is a plain-language, popular nonfiction book drawn from the author's expertise. Technical depth is used to keep the high-level statements accurate, not to turn the book into a manual.
How is it different from a vendor guide?
It is written by an operator, not a vendor. No equipment is sold or recommended, and no employer or site is named. Examples are anonymized and generalized.
Where can I buy it?
See Where to buy above.
Is there an ebook or audiobook?
Planned formats are ebook, paperback, hardcover, and audiobook. Formats will be listed in the facts table above when confirmed.
What is the Inside Hyperscale series?
Inside Hyperscale is the series name for the author's books on hyperscale data centers. Where AI Lives is Book 1 in the series.