Scale AI, the fast-rising company that helps tech firms prepare data to train their artificial intelligence models, has launched a legal battle against one of its former sales employees and competitor Mercor.
The lawsuit, filed Wednesday, alleges that the employee, Eugene Ling — who recently joined Mercor — “stole more than 100 confidential documents concerning Scale’s customer strategies and other proprietary information,” according to a copy of the suit reviewed by TechCrunch.
The case centers on claims of trade secret misappropriation against Mercor and breach of contract against Ling. Scale contends that Ling was actively pitching Mercor to one of Scale’s largest customers, referred to in the filing only as “Customer A,” before formally leaving his position at Scale.
Register for Tekedia Mini-MBA edition 19 (Feb 9 – May 2, 2026): big discounts for early bird.
Tekedia AI in Business Masterclass opens registrations.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Register for Tekedia AI Lab: From Technical Design to Deployment (next edition begins Jan 24 2026).
Mercor co-founder Surya Midha pushed back, insisting the company had no use for Scale’s confidential material. In an emailed statement to TechCrunch, he acknowledged Ling may have been in possession of some documents but said they were never accessed.
“While Mercor has hired many people who departed Scale, we have no interest in any of Scale’s trade secrets and in fact are intentionally running our business in a different way. Eugene informed us that he had old documents in a personal Google Drive, which we have never accessed and are now investigating,” Midha said.
Midha added that Mercor had contacted Scale six days earlier to propose a resolution: “We reached out to Scale six days ago offering to have Eugene destroy the files or reach a different resolution, and we are now awaiting their response.”
Still, Scale claims the materials Ling allegedly held could directly enable Mercor to serve Customer A and other top-tier clients. The company asked Mercor to provide a complete list of files in Ling’s drive and to bar him from working with Customer A. According to the filing, Mercor refused.
The complaint notes that if Mercor were to secure Customer A, the contract would be worth “millions of dollars.” Though the customer’s identity is not revealed, the implication underscores how valuable Scale considers the relationship.
The case comes against the backdrop of a fierce race in the LLM training industry, where talent poaching has become the norm. Even with Meta’s $14.3 billion investment in June for a 49% stake in Scale — which included hiring away its founder — Mercor has been gaining ground. TechCrunch has previously reported that Meta’s core AI unit, TBD Labs, continues to use Mercor and other providers despite its multibillion-dollar bet on Scale.
Mercor’s strategy is distinctive: it hires content specialists, often PhDs, to label and structure data in their areas of expertise. This approach has helped the smaller company carve out a reputation for quality in a field dominated by players with massive funding.
Scale, meanwhile, has endured some turbulence. After Meta’s investment, several of Scale’s largest customers — themselves rivals to Meta in AI — reportedly severed ties, concerned about potential conflicts of interest. That exodus has heightened the pressure on Scale to protect existing contracts and prevent rivals like Mercor from siphoning off business.
Analysts believe this lawsuit illustrates a trend likely to accelerate: as AI companies battle for scarce talent, employee poaching will increasingly raise risks of trade secret theft and confidential document leaks.
In Silicon Valley and beyond, companies are aggressively hiring engineers, researchers, and specialists from one another. That churn makes it difficult to draw clear boundaries between what knowledge workers carry in their heads versus what they might take in files or emails — a gray area that often leads to courtroom disputes.
While intellectual property battles are not new in tech, the stakes in AI are particularly high. The data pipelines and client relationships underpinning LLM training can be worth millions per contract, and losing them could tilt the balance in an industry still in its formative years.
Ultimately, the Scale–Mercor case is not just about one ex-employee’s Google Drive. Many believe it underscores how fragile competitive advantages are in the AI industry, where a single leak of confidential strategies can shift the balance of power despite billions of dollars in investment.