Scale AI’s bigger rival Surge AI seeks up to $1 billion capital raise, sources say

Surge AI, a data-labelling firm that competes with Scale AI, has hired advisors to raise as much as $1 billion in the first capital raising in the firm’s history, sources told Reuters, as it seeks to capitalise on growing user demand amid Scale AI’s recent customer exodus.

The company, founded by former Google and Meta engineer Edwin Chen, is targeting a valuation of over $15 billion, sources said, cautioning that the talks are still in early stages and the final number could be higher. The funding would be a mix of primary and secondary capital that provides liquidity for the employees.

Surge AI, which has been profitable and bootstrapped by Chen, has raked in over $1 billion in revenue last year, bigger than its better-known competitor Scale AI, which reported $870 million in revenue over the same period of time. In comparison, Scale AI was valued at $14 billion in a funding round last year, and was mostly recently valued at nearly $29 billion when Meta invested for a 49% stake in the company and poached its CEO Alexandr Wang to be its chief AI officer to lead its new Superintelligence Labs.

Surge AI declined to comment. Like other Scale AI competitors, Surge AI is benefiting from Scale AI’s customer losses following Meta’s investment. This includes OpenAI and Scale’s largest customer, Google, who are now planning to move away from the platform over concerns that doing business with Scale could expose their research priorities to Meta. Scale has said its business remains strong, and it is committed to protecting customer data.

Surge AI’s quiet yet meteoric rise has positioned it as one of the largest players in the crowded data labelling industry, defying the typical Silicon Valley playbook of raising massive rounds of venture capital to fuel growth. Founded in 2020, the San Francisco-based company has largely operated under the radar, known for its premium, high-end data labelling services used by top AI labs, including Google, OpenAI and Anthropic.

As reinforcement learning from human feedback (RLHF) has become more important in training advanced AI systems, the demand for meticulously labelled, nuanced datasets has grown. Surge AI has capitalised on this trend by appealing to a network of highly skilled contractors instead of large pools of low-wage labour.

The outsized funding of Surge would be a test of investor interest in the data labelling sector. Some investors view data labelling as an ongoing necessity for AI development, predicting a continued demand from leading AI labs. Others express concern that the industry’s low margins and reliance on human labour could make it vulnerable to automation, as AI technology advances and the need for manual annotation diminishes.