Explainer · Chapter 12 ▸ Strategy cluster

There's no AI strategy. There's strategy in an AI-shaped world.

Chegg had a perfect online-education strategy in 2020. ChatGPT killed them in 18 months. Singapore had nothing in 1960 and bet on the system shift. The strategic question isn't "what's our AI strategy." It's "where do we play, and how do we win, in a world AI has restructured."

"What's our AI strategy" is the wrong question because it frames AI as a discrete initiative. Strategy isn't about AI. Strategy is about where to play and how to win - and AI has changed both.

The firms that build AI strategies are answering the small question. The firms that rewrite their strategy in an AI-shaped landscape are answering the right one.

▍ The mechanism

Where to play and how to win, restructured

Where to play used to be a relatively stable question. Industries stayed in their lanes. Markets had decade-long structures. The choices were "which segment, which geography, which adjacency." AI is reshaping all of them simultaneously - collapsing some categories, creating others, blurring industry boundaries.

How to win used to be about extracting more from the existing playing field - better product, better customer experience, better unit economics. AI changes what "winning" even looks like, because the basis of competition shifts when an industry reshuffles.

The strategic exercise isn't "how do we add AI." It's "given how AI has restructured the landscape, which square should we play in and what's our move within it?" The answers are usually different from what your existing strategy says - and that's the painful, important conclusion.

▍ Historical analogue

Chegg vs Singapore

Two responses to a structural shift. Two completely different decades.

Chegg in 2020 had a textbook online-education business - textbook rentals, homework help, tutoring. The strategy was sound. Customer base loyal. Revenue growing.

ChatGPT launched November 2022. Within 18 months, Chegg's stock had dropped 99%. The strategy hadn't failed; the ground had moved. Homework help didn't need a Chegg subscription when an AI assistant did it free. The where-to-play assumption - "students need a paid alternative to web search for academic help" - had stopped being true.

Singapore in 1960 had nothing - no resources, no industry, just a port. Lee Kuan Yew's government bet not on extending what they had but on betting on the next system shift (containerisation, electronics manufacturing, financial services). They re-picked the playing field every 15 years as the world changed. The result: from struggling port to one of the richest economies on earth.

▍ Two strategic moves

Optimise the old game vs re-pick the field

Move A

Optimise the existing

Posture: Stay in your market; deploy AI for advantage
Confidence: High - you know this market
Risk: Your market gets restructured around you
Outcome: Optimised version of the old business; eventually obsolete

Move B

Re-pick the playing field

Posture: Treat AI as a system shift; ask which markets are now buildable
Confidence: Low at first - you're entering territory you don't know
Risk: You bet wrong on the new playing field
Outcome: Position in a market that AI made possible; durable if you got the bet right

Most leadership teams pick Move A by default because it's legible to the board and continuous with the existing business. The boards of Kodak, Sears, and Chegg all picked Move A. The boards of Apple in 2007 and Netflix in 2011 picked Move B.

▍ Same choice, two different decades

Two firms standing at the same fork

A media incumbent picking Move A

A major newspaper invests heavily in AI-augmented production - AI writes first drafts, AI summarises wire reports, AI optimises headlines for SEO. Costs drop. Output volume rises.

The where-to-play assumption - "people will pay for our news" - wasn't tested. AI assistants are increasingly the place readers ask for news summaries; the newspaper's traffic shrinks even as its production gets more efficient. They're more productive at a job the market is moving away from.

An incumbent picking Move B

A traditional retail bank re-asks "where do we play." Decides the future of consumer banking is being a trusted AI agent's preferred backend - not a customer- facing app. Re-architects its product line as APIs and embedded financial services rather than direct consumer distribution.

Painful internally. Existing consumer brand teams shrink. New B2B/API teams grow. By year five, the bank is the preferred backend for several AI-native fintechs and earns more from infrastructure than from branded consumer banking. They re-picked the playing field.

▍ Apply it

Where to play. How to win. Pick a path.

The choice unfolds over the next decade.

Your firm doesn't need an AI strategy. It needs a strategy for a world AI has restructured. You can play the Chegg game - optimise the existing business with AI added on - or the Singapore game - re-pick the playing field based on what the new world makes possible.

▍ Read these next

Connects to

Frames

First, Second, Third Order

The first-order win is small. The second and third order rewrite the industry.

Strategy

Dependence vs Dominance

Modern ecosystem control isn't ownership. It's being indispensable.

Frames

The Container Logic

Real revolutions come from coordination tech, not faster machines.

Strategy

Vertical Encroachment

Tool providers eat solution providers the moment the work standardises.

Organizations

The AI-Native Firm

Built around AI as the nervous system - not bolted on top of legacy work.

You don't need an AI strategy. You need a strategy for a world AI has restructured.

If your strategy still assumes the pre-AI playing field

AI Value Stack & Agentic Architecture

An advisory engagement that re-asks "where to play and how to win" from first principles given how AI has restructured your industry - and lands on a strategy that's defensible in the world that exists now.

Best forLeadership teams that want to move beyond AI use cases and understand how AI changes the architecture of value creation in their industry.

← Back to The Reshuffle Map