Companies Are Selling Workers' Private Messages and Emails as AI Training Data

Failed companies are discovering that their internal communications and operational data have become valuable commodities in the booming AI industry.

Forbes reports that when Shanna Johnson closed down cielo24, the transcription and captioning company she led as CEO, she found an unexpected source of revenue in what she calls the company’s “operational exhaust.” This digital residue, accumulated over 13 years of business operations, included conversations on messaging system Slack, Jira IT tickets, email correspondence, and multi-terabyte Google Drive archives documenting the daily work of her employees.

Johnson partnered with SimpleClosure, a startup specializing in company wind-downs, to handle the standard closure procedures including payroll termination, tax filings, investor consent collection, and IRS paperwork. But SimpleClosure also facilitated something novel: selling cielo24’s entire digital footprint as training material for AI systems. The sale generated hundreds of thousands of dollars for the defunct company.

This emerging practice represents more than an isolated case. It signals a new dimension in the competitive landscape of AI development. As AI companies exhaust publicly available internet content for model training, they increasingly seek alternative data sources that capture authentic workplace activities and decision-making processes.

According to former OpenAI chief scientist Ilya Sutskever, AI companies had consumed all readily available public internet content by late 2024, including Reddit discussions, Wikipedia articles, and digitized books. More importantly, such public data proves insufficient for developing agentic AI systems designed to perform actual work tasks. The detailed records of how defunct companies operated daily provide precisely the type of training material needed to build competent workplace AI agents.

Ali Ansari, whose company micro1 offers a product called Roots to AI laboratories, explained the shift in requirements. “Model companies are realizing the noise in the real-world environments is required to accurately test models,” Ansari said. Roots functions as a simulated holding company where AI agents practice skills including financial services management and complex calendar coordination.

The surge in demand for workplace data has significantly benefited SimpleClosure. CEO Dori Yona described the level of interest from AI companies as “insane” and noted “There’s a feeling of a gold rush from these companies trying to get their hands on real-world data.”

To capitalize on this demand, SimpleClosure is launching Asset Hub, a platform where closing companies can sell code repositories, Slack archives, emails, and similar materials. Parts of Asset Hub remain in beta testing because SimpleClosure removes all personally identifiable information from internal company data, a technically challenging process that Yona wants perfected before wider deployment. Over the past year, SimpleClosure has completed nearly 100 transactions on behalf of defunct companies, recovering over one million dollars for founders, with typical payments ranging from $10,000 to $100,000 per company.

However, the practice raises significant privacy concerns. Marc Rotenberg, founder of the Center for AI and Digital Policy, questions whether employers should sell internal communications to third parties, even when employees have signed intellectual property agreements covering work materials. Employees likely never anticipated their Slack messages being repurposed for AI training. “I think the privacy issues here are quite substantial,” Rotenberg said. “Employee privacy remains a key concern, particularly because people have become so dependent on these new internal messaging tools like Slack…It’s not generic data. It’s identifiable people.”

Rotenberg’s organization sent a letter to the Senate Commerce Committee calling for FTC scrutiny of new AI business practices, expressing concerns about personal data protection safeguards.

The instant bestseller Code Red: The Left, the Right, China, and the Race to Control AI, written by Breitbart News social media director Wynton Hall, serves as a blueprint for conservatives to create effective policies around AI not only for the nation, but also their family. This becomes even more crucial as newer and more powerful AI systems hit the market, and business communications once thought private are treated as input for AI’s insatiable need for raw data.

Senator Marsha Blackburn (R-TN), who was named one of TIME’s 100 Most Influential People in AI, praised Code Red as a “must-read.” She added: “Few understand our conservative fight against Big Tech as Hall does,” making him “uniquely qualified to examine how we can best utilize AI’s enormous potential, while ensuring it does not exploit kids, creators, and conservatives.” Award-winning investigative journalist and Public founder Michael Shellenberger calls Code Red “illuminating,” ”alarming,” and describes the book as “an essential conversation-starter for those hoping to subvert Big Tech’s autocratic plans before it’s too late.”