Mercor Cyberattack Linked to Compromised LiteLLM Project
▼ Summary
– Mercor, an AI recruiting startup, confirmed a security incident resulting from a supply chain attack on the open-source project LiteLLM.
– The extortion hacking group Lapsus$ claimed responsibility for targeting Mercor and leaked a sample of its data, including Slack and ticketing information.
– Mercor facilitates over $2 million in daily payouts and was valued at $10 billion after a funding round in late 2025, working with companies like OpenAI.
– The company stated it moved promptly to contain the incident and is conducting an investigation with third-party forensics experts.
– The LiteLLM compromise involved widespread malicious code, prompting the project to change its compliance processes, though the full impact remains unclear.
A significant supply chain attack involving the widely used open-source project LiteLLM has impacted the AI recruiting platform Mercor. The startup confirmed this week that it was among the thousands of organizations affected by the compromise, which has been linked to a hacking group known as TeamPCP. The disclosure follows separate claims by the notorious extortion group Lapsus$, which asserted it had targeted Mercor and gained access to its data. The precise connection between the two incidents remains under investigation.
Mercor, founded in 2023, partners with leading AI firms like OpenAI and Anthropic to train models by contracting specialized experts such as scientists and lawyers, primarily from markets like India. The company facilitates over $2 million in daily payouts and achieved a $10 billion valuation last October following a substantial funding round. In response to the security event, Mercor spokesperson Heidi Hagberg stated the company acted swiftly to contain and address the issue. “We are conducting a thorough investigation supported by leading third-party forensics experts,” Hagberg said. The company pledged to communicate directly with customers and contractors and devote necessary resources to resolution.
Prior to Mercor’s statement, Lapsus$ posted a sample of allegedly stolen data on its leak site. The sample, reviewed by journalists, included references to internal Slack communications and ticketing data, along with videos showing interactions between Mercor’s AI systems and its platform contractors. Hagberg declined to comment on the potential link to Lapsus$ or to confirm whether any customer or contractor information was accessed, exfiltrated, or misused.
The broader LiteLLM compromise came to light last week when malicious code was discovered in a package associated with its open-source library. Although the code was removed quickly, the incident raised serious concerns due to the project’s massive scale, with security firm Snyk noting it is downloaded millions of times daily. In the aftermath, LiteLLM has revised its compliance processes, shifting from one provider to another for its certifications. The full scope of the attack, including how many companies were affected and whether any data was exposed, is still being determined as investigations proceed.
(Source: TechCrunch)
