Micro1, a three-year-old startup that helps AI companies find and manage human contractors for labelling and training their data, has raised a $500 million, $35 million Series A funding round. The round was led by O1 Advisors, a venture capital firm co-founded by Dick Costolo and former CEO and COO of Twitter, Dick Costolo and Adam Bain.
Startups are one of many companies trying to fill the gap in the data market created by recent changes involving scale AI. After Meta invested $14 billion in scale AI and hired CEOs, AI Labs, including Openai and Google, said it plans to cut ties with startups over concerns that their research could possibly end in the hands of Meta. (Scale AI denies that it shares sensitive information with Meta as part of its partnership).
However, AI Labs still needs these data services, and startups like Micro1 aim to pick up Slack.
Ali Ansari, the CEO of Micro1, just 24, told TechCrunch that his company is working with leading AI labs, including Microsoft, and several Fortune 100 companies. Ansari said Micro1 is currently generating $50 million in annual revenue (ARR), up from $7 million when it launched in 2025.
This is still quite a cry from a massive competitor like Melkor. This has generated over $450 million in ARRs, reportedly bringing $1.2 billion in 2024. However, the growth and adoption of Micro1 among AI labs appears to be growing at a healthy rate.
As part of the new funding, Micro1 is adding Bain to its board, along with Joshua Browder, founder and CEO of AI Legal Assistant DonotPay.
“The only way a model actually learns is to use new human data from the net. Micro1 is in the core that provides that data to all frontier rabos, moving at speeds that have never been seen before,” Bain said in a statement from TechCrunch.
TechCrunch Events
San Francisco
|
October 27th-29th, 2025
Reuters previously reported details of Micro1’s fundraiser.
All these companies (MICRO1, SURGE, MERMSOR, and SCALE AI) have access to large bases of human contractors that can label and generate AI training data. The need for companies such as Openai, humanity, meta, and Google to build cutting-edge AI models has become an important service.
Scale AI was the first to dominate this space. The first insight is that relatively few can be done to low-skilled contractors around the world to help label data in AI model training. However, Ansari says that the demands of AI labs have been changing in recent years, and companies now need high-quality data labeling from domain experts such as senior software engineers, doctors and professional writers to improve their AI models. The difficult part is recruiting these types of people.
This has created Zara, an AI recruiter. This led to interviewing candidates who call experts, to work as one of the company’s contractors, or as called Ansari, and winning veterinarians. Micro1 says Zara is recruiting thousands of experts, including professors at Stanford and Harvard, and the company plans to add hundreds more each week.
The market for AI training data seems to be changing again. Today, many AI labs are interested in working with startups to develop “environment.” This is a virtual workspace that can be used to train AI agents on simulated tasks. Ansari says Micro1 is building new products in the environmental space to meet this demand.
Fortunately, for startups like Micro1, AI Labs appears to be working with multiple training data providers. The nature of the business is because it is difficult for one company to handle all of the data needs of one AI lab. So, at least for now, you can avoid a lot of business.
