Utkarsh Amitabh, an India-origin UK-based entrepreneur said he was not looking for a new role when 2025 began, his life was busy in multiple ways. So when a data-labeling startup reached out to him in January with an unusual offer, it wasn’t money or ambition that caught his attention – it was curiosity.
In an interview with CNBC Make It, Amitabh said he was approached by micro1, a startup that connects domain experts with companies training artificial intelligence models. At 34, he said he was already juggling multiple roles including author, university lecturer, founder and CEO of global mentorship platform Network Capital and a Ph.D. student at the University of Oxford’s Saïd Business School. On top of that, he had a newborn at home.
Still, the opportunity intrigued him. “Intellectual curiosity drew me in,” he told CNBC Make It. Training enterprise AI models felt closely aligned with his background in “business strategy, financial modeling and tech,” he added.
Earning $200 an hour – but money wasn’t the main driver
For his work training AI models, Amitabh earns $200 per hour. According to a pay stub reviewed by CNBC Make It, and confirmed by a company spokesperson, he has made close to $300,000 since January, including project completion bonuses.
Despite the impressive figure, Amitabh says compensation was not what motivated him most. “Money was less of a motivator,” he said, explaining that he already earned a livable income from his other professional roles.
Micro1 recruits experts with deep, specialised knowledge across fields ranging from medicine and law to engineering and technology. Amitabh, who describes himself as a “deep generalist,” fits neatly into that category.
Given that background, the role with micro1 felt like a “natural” extension of what he was already doing, Amitabh told CNBC Make It. The flexibility of the arrangement also mattered. “This didn’t seem like an add-on, but something that I could use to further my interests in a limited number of hours a week,” he told the financial wellness publication.
In a statement to CNBC Make It, micro1’s chief marketing officer Daniel Warner said these specialists form the “backbone of our data quality.”
The ‘trillion-dollar question’: AI vs Jobs
As AI becomes more deeply embedded in workplaces, fears about job losses continue to grow. While responding to a question – “Could helping train AI today reduce opportunities for people like him tomorrow?”
Amitabh told the publication, “This is the trillion-dollar question”. He believes most people fall into either “techno-optimist or techno-pessimist” camps when thinking about AI’s impact on work. “I like to think of myself somewhere between a techno-optimist and a techno-realist,” he added.
He acknowledges there will be “growing-up pains” as AI tools become more common, and that some jobs will inevitably disappear – a trend that HR leaders say has already begun.
At the same time, Amitabh shares the more optimistic view that AI will also create new roles. A January 2025 World Economic Forum analysis predicted that while AI will disrupt the global labor market, it could result in nearly 80 million net job gains by 2030.
Breaking down complex problems for machines
Many of the projects Amitabh works on are confidential. In general terms, he told the publication that the work involves “looking at a complex business problem that a regular user, a business owner or an executive, might have, and then breaking down that problem into small parts.”
“You need to have immense attention to detail, and you have to often look out for mistakes that the human might make or a machine might make, and you discover more about the kinds of mistakes that exist by the process of immersing yourself in it,” he said,
Amitabh describes the work as “intellectually quite demanding,” especially because AI models are constantly improving. That evolution forces even seasoned experts to continuously upgrade their own knowledge and creative thinking.
Still, he finds the end goal motivating. “The ultimate goal is actually really energizing,” he said. “You’re seeing whether the machine and human, the way this engagement is happening, [can] level up the output for problems that you asked and other kinds of problems that might be related to it.”