The Best Books on AI Strategy: A Reading List That Actually Holds Up

Most AI strategy reading lists are really just AI hype lists. They tell you AI is transforming everything, that leaders must adapt, and that the future belongs to the bold. You finish the book feeling energized and no more capable of making a strategic decision than you were before you started.

These books are different. Each one gives you a framework, a mechanism, or a mental model that changes how you think about a specific problem. Some are about AI directly. One isn't about AI at all—and it might be the most important one here.

This list is weighted toward strategy vs tactics in marketing: not which tools to use, but how to decide where AI changes the structure of competition, the distribution of value, and the decisions your organization needs to make differently. Read them in roughly this order.

---

The List

Never Always, Never Never: Strategic Marketing in an AI World by Patrick Gilbert

Published in 2026 by DELLROCK MEDIA, Never Always, Never Never opens with a premise most AI books avoid: many organizations are using AI to accelerate the wrong work. The core argument is that strategy should return to the center of marketing, and that AI changes who can be effective at it. Not by replacing expertise, but by giving it reach it never had before.

Across 33 chapters, the book works through why the old playbook stopped working and what AI actually changes about marketing effectiveness. One of the book's sharpest observations is the distinction between AI as efficiency tool and AI as a source of new value. Patrick Gilbert describes this using a double helix framework. The first strand is using AI to do existing work faster. The second is using it to build capabilities that weren't previously accessible. Most organizations stop at the first strand and call it transformation.

The book is also unusually honest about the gap between AI demos and AI reliability. Gilbert recounts the internal experience at AdVenture Media of building an AI-powered strategy tool that was fluent, confident, and occasionally wrong in ways that were hard to catch. That problem dissolved only when the team separated deterministic data infrastructure from the AI layer sitting on top of it. That distinction, between where probabilistic AI belongs and where it doesn't, is worth the price of the book on its own.

For marketers specifically, the chapters on how AI changes the relationship between strategy and execution are some of the clearest writing on the subject available. If you work in AI marketing strategy and you read one book from this list, make it this one. See also the companion AI Double Helix framework for a practical breakdown of the two-strand model.

---

Power and Prediction: The Disruptive Economics of Artificial Intelligence by Ajay Agrawal, Joshua Gans, and Avi Goldfarb

Released in 2022, Power and Prediction is the most consistently recommended AI strategy book in business-oriented reading lists right now, and for good reason. Agrawal, Gans, and Goldfarb extend the framework from their earlier Prediction Machines to explain something more consequential: when AI lowers the cost of prediction, it doesn't just create efficiency. It redistributes power inside organizations and across markets.

At its core, the book argues that AI disrupts the structure of decisions. Whoever controls the prediction controls the decision, and whoever controls the decision accumulates power. That's a different conversation than productivity, and it's the one most executive teams are not having. It carries a Goodreads rating of 4.22, which is unusually high for a business strategy book of this density.

---

Competing in the Age of AI: Strategy and Leadership When Algorithms and Networks Run the World by Marco Iansiti and Karim R. Lakhani

From Harvard Business Review Press in 2020, this book makes a structural argument: AI isn't a feature you add to an existing business. It's a new operating foundation that requires rearchitecting how you create and deliver value. Iansiti and Lakhani focus on how algorithms and network effects interact to create winner-take-most dynamics, and why layering AI onto legacy processes produces far less value than redesigning those processes around AI from the start.

One fair criticism is that the book is more theoretical than prescriptive. It diagnoses the problem with clarity but is less useful as an implementation guide. Read it before you finalize your operating model, not after.

---

Good Strategy Bad Strategy: The Difference and Why It Matters by Richard P. Rumelt

Written in 2011, this is not an AI book. It belongs on this list anyway.

Rumelt's argument is deceptively simple: most things called strategies are not strategies. They are goals dressed up as plans, or ambitions disguised as direction. Real strategy, he argues, has a kernel. A diagnosis of the core challenge, a guiding policy that addresses it, and coherent actions that reinforce each other. Everything else is noise.

What makes this book essential for AI strategy specifically is that most AI strategy discussions fail at this first step. Organizations announce AI initiatives without diagnosing what the actual challenge is. They adopt tools before they've chosen a direction. The result is activity without progress. Rumelt gives you the discipline to avoid that trap. Read it first, or read it alongside everything else on this list as a corrective.

---

Prediction Machines: The Simple Economics of Artificial Intelligence by Ajay Agrawal, Joshua Gans, and Avi Goldfarb

Published in 2018, Prediction Machines is the book that gave executives a clean economic lens for thinking about AI before the current wave of hype arrived. The central framing is elegant: AI is a technology that reduces the cost of prediction. That's it. And when the cost of prediction falls, the value of complementary inputs, judgment, data, action, rises.

Still the clearest framework for deciding where AI creates genuine economic advantage versus where it's just substituting for something you could do another way. Read Prediction Machines before Power and Prediction. They build directly on each other.

---

Books on Adoption and Working Methods

Co-Intelligence: Living and Working with AI by Ethan Mollick

Out in 2024, Co-Intelligence is the most frequently recommended starting point for business readers in current reading lists. Ethan Mollick focuses on how individuals and organizations can actually work with AI. Not the theory of transformation, but the practice of collaboration with tools that are powerful, imperfect, and constantly changing.

It leans more toward adoption and working methods than pure strategy, which is why it's lower on this list. But if your team is still in the phase of figuring out how to build an AI-first culture, how to get people using these tools seriously rather than superficially, Co-Intelligence is the right place to start. Mollick's emphasis on experimentation over perfection maps directly to what makes AI adoption actually work in practice.

---

Books on Human-AI Collaboration

Human + Machine: Reimagining Work in the Age of AI by Paul R. Daugherty and H. James Wilson

First published in 2018, Human + Machine argues that competitive advantage comes from redesigning work so humans and AI complement each other, particularly in decision support and scaled operations. The "hybrid intelligence" framing was prescient in 2018 and reads as obvious now, which is either a criticism of the book or proof of how right it was.

It's more an enterprise transformation book than a sharp strategy text, and it has been somewhat overtaken by newer work. Worth reading if you're thinking about how to build an AI-first marketing team and want the foundational thinking on human-AI collaboration before the current tools existed.

---

Books on Macro Stakes and Governance

The Coming Wave: Technology, Power, and the Twenty-First Century's Greatest Dilemma by Mustafa Suleyman and Michael Bhaskar

Released in 2023, The Coming Wave pulls back further than any other book on this list. Mustafa Suleyman argues that AI is part of a broader wave of rapidly scaling technologies that will require new governance, institutional responses, and strategic positioning at the macro level.

This is the book for understanding the structural stakes, not the operational playbook. It's less useful for deciding what your company should do in the next 18 months, and more useful for understanding why the decisions you make in the next 18 months will matter for the decade after that. Read it when you have the bandwidth to think at that scale.

---

How to Use This List

Don't read these in parallel. The books build on each other in ways that matter.

Start with Rumelt if you've never had a rigorous framework for what strategy actually is. Then move to Prediction Machines for the economic foundation, followed by Power and Prediction for the implications. Competing in the Age of AI gives you the operating model implications. Co-Intelligence brings it back to the practical level of how people and teams actually adopt this.

Never Always, Never Never sits at the intersection of marketing and AI strategy specifically. It assumes you already understand the basics of how brands grow and builds from there, which is why it works well either first or last depending on your background.

As you read, keep returning to one question: where does AI actually change the structure of your decisions, not just the speed of your execution? That's what these books, taken together, are trying to answer. For more on how strategy vs tactics shapes AI adoption, and how AI changes the relationship between brand and performance marketing, the learn section of this site covers the frameworks in depth.

For related reading, see our posts on the AI maturity ladder and how to become an AI-first organization.