Founders of defunct startups are discovering a new revenue stream by selling their accumulated data to AI labs training and other AI systems. One former crypto startup founder earned roughly $25,000 selling his company's codebase and data archives after deciding to wind down the business. The trend reflects growing demand from AI companies seeking diverse datasets to train their models, and offers a lifeline to founders who might otherwise lose money when their ventures fail.
This emerging market represents a shift in how startup failure is monetized. Rather than data disappearing when companies shut down, AI labs are actively acquiring these digital assets at prices founders consider meaningful. The practice raises questions about data provenance, user privacy, and what happens to the terms of service users agreed to when the original company was operating.
This is an AI-written summary. The original reporting belongs to MIT Technology Review. How we write these · Spot a mistake?