The web development education ecosystem — independent course creators, technical authors, documentation writers, DevRel professionals — is collapsing. Not slowly, not ambiguously. Baldur Bjarnason's compilation of first-person accounts from across the field documents a synchronized economic wipeout driven by generative AI's consumption of freely shared knowledge without compensation or consent. The numbers are stark. Axel Rauschmayer, one of the world's most prolific JavaScript authors since 2005, saw book income drop from livable to zero between 2024 and 2026. His blog traffic increased, but virtually all of it comes from AI crawlers generating no ad revenue — forcing him to take his work offline entirely. Josh W. Comeau, who builds interactive web development courses, reports revenue down more than 50%. YouTube creator Kyle Cook of Web Dev Simplified says his programming tutorial income has halved in a single year. These are not marginal players. They are the people who built the knowledge base that LLMs now regurgitate. The extraction mechanism is precise: AI companies scrape freely published educational content, train models on it, then sell access to that aggregated knowledge back to the same developers who previously consumed it directly from creators. The original authors receive nothing. Worse, the traffic their content generates now serves AI crawlers rather than human learners, destroying the ad-supported and course-sale business models simultaneously. Rauschmayer's situation is the purest illustration — more traffic, zero revenue, because the traffic isn't human. Rachel Andrew, a renowned technical editor, identifies a second-order extraction: AI doesn't just consume existing knowledge, it degrades the production pipeline. Writers using AI tools introduce subtle inaccuracies that editors must catch, shifting labor costs downstream while creating an illusion of individual productivity. "You might feel more productive, but all you've done is move the work around, make someone else's job or experience measurably worse, and reduce quality," Andrew writes. The productivity gains from AI are partly real and partly an accounting trick — externalizing quality costs onto editors, readers, and the broader information ecosystem. Salma Alam-Naylor's account reveals the human cost beyond economics. Developer Relations was already a role under constant pressure to justify its existence in short-term financial terms. GenAI provided the final push: developers stopped gathering in learning communities and switched to chatbots for education. Alam-Naylor left public-facing work entirely for a lower-paid developer job. The field is losing not just content but the people who know how to teach — a capability that takes years to develop and cannot be easily reconstituted. Bjarnason explicitly rejects the "adapt or die" response. He notes the cruel irony of GenAI promoters using the word "democratize" when the actual effect is the opposite: centralizing knowledge into corporate platforms, commodifying understanding into token-prediction outputs, and destroying the volunteer-driven, non-commercial documentation projects that genuinely democratized web publishing. The independent educator ecosystem that helped local businesses, schools, sports clubs, and community groups build their own web presence is being replaced by a system that concentrates value in a handful of AI companies. The structural question Andrew raises deserves serious attention: if AI's productivity gains are modest, partially illusory, and achievable through non-AI automation — while the environmental, financial, and human costs are substantial — then the current arrangement is not an efficiency improvement. It is a wealth transfer from a distributed network of knowledge workers to a concentrated set of platform companies, with no compensation mechanism and no feedback loop that would correct the imbalance.
