AI Labs Are Buying Data from Dead Startups to Train Their Models

When Keelan Miskell began winding down his crypto startup this year, he initially didn’t believe its data held much value, given its anonymous users and short operational history. However, he has since successfully monetized the firm’s digital archives. Miskell netted approximately $25,000 for his code base and anticipates an additional $15,000 for workspace data from platforms like Google, Notion, Figma, and Slack. "At least all that stuff we were doing for two and a half years is worth something to someone somewhere," he remarked. This sale was facilitated by SimpleClosure, a company at the forefront of a burgeoning sector: businesses dedicated to assisting failing startups. Estimates from Carta and AngelList indicate a significant rise in startup shutdowns, ranging from 26% to 56% annually in recent years. This trend has fostered an "end-of-life ecosystem" that includes coaches, "death doulas," and firms specializing in the legal, tax, and operational complexities of business closure. Now, these "tech undertakers" are channeling the data remnants to AI laboratories. SimpleClosure integrated data sales into its services in April, launching its Asset Hub platform to help founders sell their code and workspace data—including emails, chats, and documents. Buyers span foundation AI labs, vendors creating simulated workspaces for AI training, and established tech companies developing their own proprietary models. Asset Hub has already managed over 150 deals and projects returning more than $10 million to startups by the end of 2026. Dori Yona, CEO of SimpleClosure, notes that these transactions, once rare and "somewhat taboo," have become increasingly common, with 47% of SimpleClosure's deals now involving code or data sales. SimpleClosure is not alone in this space. Industry competitor Sunset also boasts a rapidly expanding data resale operation. Brendan Mahony, Sunset's Founder and CEO, explains that his company assesses a dying client’s assets based on factors such as head count, years in operation, data volume, and how the team utilized its tools. Mahony draws an analogy to particle board: "This operational exhaust data is like the wood shavings that everyone thought had no value, but it actually has a ton." The demand for data from defunct companies continues to grow.