MICRO1, a three -year starting start that helps AI companies find and manage human contractors for labeling and data training, has increased a $ 35 million AD funding round of $ 500 million. The round was driven by the 01 Advisors, a business capital company co -founded by Dick Costolo and Adam Bain, former CEO and Twitter COO.
Startup is one of the many companies that want to fill the gap in the data market created by recent changes that include the AI scale. After Meta invested $ 14 billion on a AI scale and hired its CEO, AI Labs, including Openai and Google, said they were planning to reduce the bonds by starting, possibly above concerns that their research could end up with META. (Scale ai refuses that she shares confidential information with Meta as part of her collaboration.)
However, AI Labs still need these data services and newly established businesses such as micro1 aim to get the loose.
Micro1 Ali Ansari – who is only 24 years old – tells TechCrunch that his company is working with top AI Labs, including Microsoft, as well as several Fortune 100 companies. 2025.
This is still far from larger competitors such as Mercor, which produces more than $ 450 million in ARR, and Surge allegedly carrying $ 1.2 billion $ 2024. However, the growth and adoption of the micro1 among AI laboratories appear to be rising at a healthy pace.
As part of the new funding, the Micro1 also adds to Bain to his board, along with Joshua Browder, founder and managing director of the legal assistant AI Donotpay.
“Really the only way models are learning now is through clean new human data. The micro1 is at the core of providing these data to all Labs Frontier, while moving at speeds I have never seen before,” Bain said in a statement to Techcrunch.
TechCrunch event
Francisco
|
27-29 October 2025
Reuters in the past reported details Micro1’s capital assembly attempts.
All these companies – Micro1, Surge, Mercor and Scale AI – provide AI workshops with access to a large basis of human contractors who can highlight and create AI training data. It has become a crucial service that companies such as Openai, Anthropic, Meta and Google have to build AI models.
The AI scale was the first to dominate this area, with the original picture that it could pay relatively few for low -skilled contractors worldwide to help highlight data on AI models. However, Ansari says the requirements of AI laboratories have been shifted in recent years and that companies are now needed high quality data from sector experts-such as superior software engineers, doctors and professional writers-to improve AI models. The hard part was made recruiting these types of peoples.
This has led Micro1 to build AI’s recruitment, Zara, who interviews and the candidates who apply to work as one of the company’s contractors or as Ansari calls on experts. Micro1 says Zara has recruited thousands of experts – including Stanford and Harvard teachers – and that the company plans to add hundreds every week.
The AI training data market seems to change again. Now, many AI laboratories are interested in working with the newly established companies for the development of “environment” – virtual workplaces that can be used to train AI agents in simulated tasks. Ansari says Micro1 is making new offers in the environment to meet this demand.
Fortunately for newly established businesses such as Micro1, AI Labs seem to work with multiple training providers. The nature of the business is such that it is difficult for any company to handle all the data needs of an AI laboratory. This means that there is enough work to go around, at least for now.
