AI data startup Micro1 reaches $500M gross run rate amid AI training boom
TechCrunch Marina Temkin
Micro1 says its annual run rate jumped from $100M to $500M in eight months. The AI training-data boom is getting big enough to feed several startups, not just one.
Based on reporting by TechCrunch, Marina Temkin — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
The rush for custom AI training data is turning into a very real business, and Micro1 is one of the clearest signs. The four-year-old startup has taken its gross annual run rate from $100 million to $500 million in just eight months, according to a person familiar with the company.
That headline number is not what Micro1 keeps. Like other firms in the same niche, it hires domain experts on contract — doctors, lawyers, scientists — and retains roughly 60% to 70% of the gross. That puts its net annual run rate somewhere between $150 million and $200 million.
Micro1 still sits behind the biggest names in the space. Mercor hit $2 billion in gross annualized revenue this summer, and Handshake reached $1 billion earlier this year. But the point is less about who is in front and more about how much room there is. The source describes demand from top labs and corporations as near-bottomless, and some researchers think AI spending on data could eventually rival spending on compute.
The company is also changing how it makes this data. It is increasingly generating synthetic material without humans in the loop, including automated descriptions of video content. Some of that output can be sold to multiple customers, and a person familiar with Micro1’s finances said those off-the-shelf datasets can carry gross margins of 80% to 90%.
That model has already stirred controversy. Critics say selling the same data to multiple clients, including Chinese AI developers, helps those systems catch up to top U.S. models. Micro1’s founder, Ali Ansari, said last month on X that the company does not sell to Chinese model makers, and he took aim at competitors that do.
Micro1 started out as an AI recruiting startup, then pivoted after seeing data-labeling customers use its platform to hire engineers for annotation. The company raised a Series A at a $500 million valuation last September, and TechCrunch understands it may have raised another round since then at a much higher valuation. The money in this corner of AI is moving fast, and the arguments around who gets the data are moving just as fast.
My take — AI-written commentary, not fact-checked reporting
This is the part of AI nobody at the keynote wants to talk about: the gold rush is not just models, it’s the people and data feeding them. The open-versus-closed fight looks quaint when the real moat is a warehouse of labeled human judgment with 90% margins and plenty of moral footnotes.
Read more about this at: TechCrunch
