Share This Article
As artificial intelligence moves beyond basic automation, young Indian-origin entrepreneurs like Surya Midha are participating in a growing industry focused on bringing human expertise into the development of intelligent systems.
Artificial intelligence is entering a new phase of development, one in which computing power and enormous datasets are no longer the only factors determining the capabilities of advanced systems. Increasingly, the industry is turning to human expertise to improve how AI models reason, interpret complex information, and perform specialized tasks.
This shift is creating opportunities for professionals across medicine, finance, law, engineering, and scientific research. It is also attracting a new generation of entrepreneurs, including Indian-origin technology founder Surya Midha, who is involved in connecting specialized human knowledge with artificial intelligence development.
Midha, a co-founder of Silicon Valley-based technology company Mercor, is part of an emerging business ecosystem supporting the development of advanced AI models. His work illustrates a broader trend in which human intelligence is becoming an important resource for improving machine intelligence.
Why Artificial Intelligence Still Needs Human Experts
Large language models have demonstrated considerable progress in generating text, writing software, and responding to complex questions. However, their ability to produce convincing answers does not necessarily guarantee accuracy or sound judgment.
In fields such as healthcare and law, a response that appears reasonable may still contain important errors. This creates a need for qualified professionals who can evaluate outputs, identify mistakes, and provide feedback based on established knowledge.
AI developers increasingly rely on subject-matter experts to help create training examples, assess model responses, and design evaluations that measure how reliably systems handle difficult tasks.
Unlike traditional data-labeling work, which often involves categorizing images or identifying objects, advanced AI training can require substantial professional knowledge. A medical specialist might assess clinical reasoning, while a financial professional could evaluate how a model approaches a complicated analytical problem.
The Growing Market for Human Expertise
The increasing demand for expert feedback has created a specialized market connecting AI developers with professionals capable of supporting model training and evaluation.
Technology companies are experimenting with ways to identify qualified contributors and incorporate their expertise into structured development processes.
Midha’s entrepreneurial work has intersected with this expanding market. Alongside Brendan Foody and Adarsh Hiremath, he established Mercor in 2023. The business initially focused on AI-assisted recruitment before expanding into providing specialized professionals for AI training.
The development reflects a wider transformation in the technology industry: as AI systems become more sophisticated, improving their performance can require increasingly sophisticated human input.
From Automation to Collaborative Intelligence
Much of the public discussion surrounding artificial intelligence has focused on whether machines will replace human workers. Yet the development of advanced AI systems also reveals another relationship between humans and technology.
Professionals are not only using AI tools to complete tasks. In some cases, they are helping determine how those tools should approach problems, assess evidence, and recognize their limitations.
This relationship is particularly important in industries where professional judgment cannot be reduced to a straightforward set of instructions.
A model trained to generate legal documents, for instance, may benefit from expert feedback about legal reasoning and jurisdictional differences. Similarly, models designed for scientific applications require careful evaluation to distinguish plausible conclusions from well-supported findings.
Human feedback does not eliminate the risks associated with AI, but it can contribute to more rigorous development and testing.
A New Generation of AI Entrepreneurs
The growth of specialized AI services has attracted founders from backgrounds that differ from those traditionally associated with major technology companies.
Midha, who has Indian family roots and grew up in California, represents this emerging generation of entrepreneurs. His involvement in the AI industry began at a young age, alongside co-founders with whom he had longstanding personal connections.
Their business gained significant attention as investment in artificial intelligence infrastructure and services accelerated. For Midha, the experience has placed him within a technology sector undergoing rapid changes in how businesses develop and deploy intelligent systems.
The wider significance extends beyond individual companies. Entrepreneurs working in AI recruitment, model evaluation, and training infrastructure are addressing problems that become more important as organizations attempt to use artificial intelligence in increasingly specialized environments.
What Comes Next for AI Development?
The next stage of artificial intelligence will likely involve more than building larger models. Reliability, domain-specific performance, and effective evaluation are becoming increasingly important considerations for developers and businesses.
Human expertise is expected to remain valuable in areas where models must interpret complex situations, follow professional standards, or produce results that can withstand independent scrutiny.
At the same time, the industry faces questions about the quality of training data, the consistency of human feedback, and how to evaluate increasingly capable systems.
The work of entrepreneurs such as Surya Midha illustrates how an emerging part of the AI economy is developing around these challenges.
As artificial intelligence advances, the relationship between human knowledge and machine capabilities may prove as consequential as advances in computing infrastructure or model architecture. The challenge for the industry will be to translate specialized human expertise into AI systems that are more capable, dependable, and useful in the real world.