Research: Vibe coding AI threatens open source ecosystem
A new economics paper indicates that generative AI, specifically “vibe coding,” may erode the open-source software ecosystem. The paper, titled “Vibe Coding Kills Open Source,” was authored by researchers from Central European University and the Kiel Institute for the World Economy.
Vibe coding involves generative AI agents assembling applications by selecting and integrating open-source components. This process often occurs without users reading documentation, filing issues, or contributing to the projects.
The authors suggest this intermediary layer can weaken engagement, through which many maintainers earn returns. This weakening could potentially shrink the supply and quality of open-source building blocks that the industry relies upon.
The paper describes vibe coding as having two opposing effects: productivity increases due to AI lowering the cost of using and building on existing code, while maintainer incentives decrease as user attention and feedback are diverted to AI interfaces.
The model predicts that if open source is primarily “monetized” through direct user engagement, increased vibe coding adoption will reduce entry and sharing, decrease variety and average quality, and could lower overall welfare, even as coding speed improves.
The paper argues that the open-source ecosystem’s incentive structure was designed for a pre-AI workflow, where users regularly interacted with maintainers via documentation, issue trackers, and community forums.
Evidence consistent with the “disintermediation” of communities is cited. Independent analyses have shown a decline in Stack Overflow posting activity, as developers use private AI chats for troubleshooting and code generation instead of public Q&A. The paper also mentions maintainer reports of increased usage coinciding with decreased documentation traffic, citing Tailwind CSS as an example where reduced documentation visits have been noted despite wider adoption.
The researchers propose a new benefit-sharing layer. Platforms could track invoked open-source projects and distribute revenue to maintainers proportionally, similar to Spotify’s payout model for music streaming. Other suggested solutions include foundation support, corporate sponsorship, and public funding, aiming to treat open source as digital infrastructure with ongoing costs rather than “free inputs.”