Extending Christensen's Framework to AI: General Purpose Disruptive Technology

If you look at AI through Christensen's disruptive innovation framework, the alignment is hard to miss. At the same time, because AI is being adopted across many industries at once, it looks like what economists call a "general purpose technology."

1. Disruption begins in new or niche markets

In the early 2010s, Robbie Allen's Automated Insights powered Yahoo's fantasy football summaries. Millions of personalized recaps at a scale no newsroom could match. The technology wasn't replacing writers one to one. It was creating entertainment news content that hadn't been economically viable before.

2. Early performance only has to be "good enough" for some users

When ChatGPT arrived with a free, simple interface, adoption started with college students and knowledge workers using AI for completing homework, summarizing readings, drafting emails, and iterating on ideas. These tasks don't require perfect accuracy. The bar is "good enough to get the job done," consistent with Christensen's early-stage foothold.

3. The technology improves and moves up market

AI has moved quickly into higher value work, including coding, analytics, legal drafting, and now strategy consulting. As a recent WSJ piece asked "if AI can analyze information, crunch data and deliver a slick PowerPoint deck within seconds, how does the biggest name in consulting stay relevant?" (see AI Is Coming for the Consultants. Inside McKinsey, 'This Is Existential.').

4. Incumbents respond based on existing business models and technology

Incumbents face a structural problem: their revenue models and cost structures weren't designed for AI, and retrofitting AI means cannibalizing what already works. For instance, even Google, which produced foundational AI research, hesitated to fully productize that research because it threatened Google's search and advertising model. Other incumbents continue to struggle.

5. New entrants and new value networks emerge

Alongside companies like OpenAI, Anthropic, and DeepSeek, there are now thousands of AI-focused startups worldwide. These startups are designing revenue models and their cost structures around AI from the start.

New Idea: AI as a General Purpose Disruptive Technology (GPDT)

Economists have long identified "general purpose technologies" like electricity and computing that transform multiple industries. But Christensen's disruption framework has typically been applied to single markets.

A GPDT combines both: a technology that follows the classic disruptive pattern across many industries at once.

Some AI startups are creating entirely new solutions. Others will displace incumbents with AI-native products that deliver more value at lower cost.

The Bottom Line

AI fits Christensen's framework, but the cross-industry scale is new. Same playbook. Much bigger stage.