AI Data Labeling Startups Handshake and Mercor Both Hit $1 Billion in Annualized Revenue

Two AI data labeling startups — Handshake and Mercor — have both crossed the $1 billion annualized revenue mark in 2026, according to The Information. Handshake's gross annualized revenue surged from $550 million in January to $1 billion today, while Mercor also hit the $1 billion pace earlier this year. The milestone reflects explosive demand for human-generated training data as AI labs race to build more capable models.

What Handshake and Mercor Do

Handshake and Mercor are both platforms that connect AI companies with human contractors who perform data labeling, annotation, and evaluation tasks. These tasks — identifying objects in images, rating AI responses, transcribing audio, and more — are the raw fuel that powers supervised learning and reinforcement learning from human feedback (RLHF). As AI labs push to improve model quality, demand for high-quality human-labeled data has exploded.

Why Revenue Is Surging

The dramatic revenue growth at both companies reflects the insatiable appetite of frontier AI labs for training data. OpenAI, Anthropic, Google DeepMind, Meta, and dozens of smaller model developers are all competing to assemble the best training datasets. As models grow more capable, the bar for useful training data rises — labs need more nuanced, expert-level annotations from domain specialists, which commands higher prices per task and drives revenue up faster than volume alone would suggest.

The Human-in-the-Loop Economy

The success of Handshake and Mercor challenges the narrative that AI will quickly replace human labor across the board. In fact, AI development currently requires vast amounts of human judgment to function. The data labeling industry — once associated with low-wage crowdwork — is evolving toward a model where specialized, high-skill annotators earn meaningful wages for expert contributions. Medical, legal, and scientific labeling tasks increasingly require credentialed professionals.

Competition and Consolidation Ahead

Despite their growth, Handshake and Mercor operate in a competitive market alongside Scale AI, Appen, and dozens of niche providers. Scale AI, backed by major venture capital and government contracts, is the market leader by revenue. But the rapid growth of newer platforms suggests the market is expanding fast enough for multiple large players to coexist — at least for now. Consolidation through acquisition is a likely outcome as AI labs seek more reliable, integrated data supply chains.

The Bottom Line

Handshake and Mercor crossing the $1 billion annualized revenue mark is a clear signal that demand for AI training data is accelerating, not plateauing. As long as frontier AI development requires human-in-the-loop supervision, data labeling will remain one of the most consistently growing segments of the AI economy — and the startups that can reliably deliver quality at scale will be among the most valuable players in the ecosystem.