Job portal Handshake shifts to AI training data, growing from tens of millions to nearly $1 billion in a year

ME News Report, April 14 (UTC+8), its annualized gross revenue from AI training business has approached $1 billion, up from $550 million in January of this year, and only $5-10 million a year ago. After deducting payments to contractors, the net income from AI training business is close to $300 million. Handshake’s existing university recruiting software business has an additional annualized gross revenue of about $150 million. Handshake is at least the fourth data annotation company in recent years to surpass $1 billion in annualized gross revenue. Previously, Surge AI reached this scale first in 2024, and Scale AI also achieved it before Meta acquired nearly half of its shares. The three-year-old AI outsourcing platform Mercor also broke through $1 billion earlier this year (last September it was $500 million), with net income of $300-400 million after contractor expenses, already achieving free cash flow profitability. However, after a data breach at the end of March, Meta has suspended cooperation with it indefinitely, and it is unclear whether revenue has been impacted. Among other competitors, Turing’s annualized revenue reached $300 million by the end of 2024, and AfterQuery announced last week that it surpassed $100 million, with a valuation of $300 million. These companies’ clients include AI labs such as Anthropic and OpenAI, whose core business is organizing lawyers, PhDs, doctors, and other professionals to evaluate models and answer questions in specialized fields on a contract basis. The demand for high-quality training data produced by domain experts from AI labs continues to expand, but the moat of this business is limited: leading labs can easily switch between suppliers, and some companies choose to build their own teams. xAI, for example, directly hired over 1,000 remote trainers. (Source: BlockBeats)