AI training startup Mercor has raised $750 million in new funding. This brings the company to a $20 billion valuation as demand grows for technologies that help artificial intelligence systems perform complex professional tasks.
The investment marks another major milestone for the San Francisco-based company. Mercor has moved beyond its origins in AI-powered recruitment to become a provider of specialized human expertise for AI model development.
Mercor is now targeting a much larger opportunity: helping businesses train, evaluate, and deploy AI agents capable of handling real workplace responsibilities. Furthermore, its latest financing reflects growing investor confidence in the infrastructure needed to make those systems reliable.
Mercor Secures $750 Million as Valuation Doubles
Mercor’s latest Series D funding round values the company at $20 billion, double its reported $10 billion valuation in October 2025. The financing attracted major technology and investment firms. These include Benchmark, Felicis, General Catalyst, NVIDIA, Salesforce, Sapphire Ventures, and Zoom. There were also several other institutional investors.
The funding comes at a time when AI developers are spending heavily on improving model performance beyond traditional benchmarks. Investors are increasingly looking at companies that provide the data, expertise, and evaluation systems required to make AI useful in commercial environments. In fact, Mercor has positioned itself directly in that market. It offers businesses access to specialized professionals and training infrastructure that can help AI systems handle increasingly sophisticated assignments.
Mercor Expands Beyond AI Recruitment Into Model Training
Mercor originally built its business around using artificial intelligence to match companies with qualified professionals. Over time, that recruitment model gradually evolved into a platform connecting AI developers with experts who can contribute to model training, testing, and evaluation.
The company now works with professionals across fields such as finance, law, healthcare, and technology. Their knowledge helps AI developers understand how complex work is actually performed. This includes the decisions, judgment, and specialized reasoning required in professional settings. This approach addresses a persistent limitation in AI development. Models can produce convincing answers without necessarily meeting the accuracy or quality standards expected by industry specialists.
Mercor’s network of more than five million experts gives the company access to a broad pool of professional knowledge. That network has become a central part of its business strategy. AI companies seek more sophisticated ways to improve model performance, and Mercor supports that goal.
AI Agents Become a Major Focus of Mercor’s Expansion
Mercor is placing greater emphasis on AI agents, systems designed to carry out tasks and interact with digital tools with varying degrees of independence. Unlike conventional chatbots that primarily generate responses, AI agents can potentially complete multi-step assignments. These include research, analysis, documentation, and software applications.
The difficulty lies in making those agents dependable. For example, a model that performs well during a demonstration may struggle when faced with unfamiliar business procedures, complicated instructions, or industry-specific requirements. Mercor wants to address that problem by developing training environments and evaluations. These will reflect actual workplace conditions.
The company has also expanded its capabilities through its acquisition of Deeptune, a startup focused on simulated enterprise software environments. These simulations allow AI systems to practice tasks within representations of business applications before companies deploy them in operational settings. For Mercor, that technology provides another way to connect model training with the practical demands of enterprise customers.
Human Expertise Remains Central to Advanced AI Development
Despite rapid improvements in artificial intelligence, professional knowledge remains an important part of training systems to perform specialized work. Mercor’s business depends on that relationship between human expertise and machine learning. This is especially important as developers attempt to improve AI performance in fields where mistakes can have serious consequences.
Financial analysis, legal research, and medical documentation require more than producing fluent text. They involve interpreting information, applying established standards, and recognizing situations where additional judgment is necessary. Professionals working through Mercor can help developers identify those requirements and evaluate whether AI-generated outputs meet them.
This reliance on human expertise also raises questions about the future of professional employment. Experts are helping build systems that may eventually automate portions of their own work. For now, however, demand for specialized knowledge has created a commercial opportunity for platforms that connect qualified professionals with AI developers.
Mercor’s Growing AI Spending Signals a Shift in Business Operations
Mercor’s investment in artificial intelligence extends beyond the services it sells to customers. Co-founder Brendan Foody has disclosed that the company spends approximately three times as much on AI inference as it does on employee salaries. This illustrates how extensively AI computing has become integrated into its operations.
Inference refers to the computing resources used when trained AI models process requests and generate results. As companies deploy more sophisticated AI systems, these ongoing operational costs can become substantial. Mercor’s spending reflects a business model in which computing infrastructure plays an increasingly prominent role alongside human employees.
That approach may offer a glimpse of how certain technology companies will operate in the coming years. Rather than expanding their workforce at the same pace as revenue, businesses may invest more heavily in AI systems that support employees and automate repeatable processes. Whether those investments consistently produce the expected productivity gains remains an important question.
Enterprises Face Pressure to Deliver Returns From AI Investments
Mercor’s funding arrives as businesses increasingly focus on the financial returns generated by artificial intelligence. Many organizations have already experimented with generative AI tools. However, turning those experiments into reliable systems capable of handling business operations remains difficult.
Research cited by PYMNTS Intelligence found that 64% of AI leaders reported very positive returns from their AI investments. At the same time, 66% expected it would take at least three years to realize the full benefits of those investments. The findings suggest that enthusiasm for AI remains strong even as companies acknowledge the time required to integrate the technology effectively.
This gap between AI adoption and measurable business results creates opportunities for companies such as Mercor. Enterprises need ways to evaluate whether AI systems can perform specific assignments accurately, follow internal procedures, and operate within acceptable cost limits. Therefore, training models against real business requirements could become an increasingly important part of enterprise AI development.
What Mercor’s $20 Billion Valuation Means for the AI Industry
Mercor’s $20 billion valuation reflects a broader shift in the artificial intelligence market. Investors are no longer concentrating exclusively on companies developing large language models or manufacturing the hardware needed to run them. Businesses providing training infrastructure, expert-generated data, and evaluation technologies are also attracting substantial capital.
The opportunity is particularly significant as AI development moves toward systems capable of performing work rather than simply answering questions. Building those systems requires more than powerful models. Developers must understand how tasks are completed, establish reliable performance measurements, and identify where automated systems fail.
Mercor is betting that its combination of human expertise, AI training infrastructure, and enterprise evaluation tools will become increasingly valuable as companies adopt AI agents.
Its latest funding gives the startup more resources to pursue that ambition. The bigger challenge will be demonstrating that its technology can help businesses deploy AI systems that deliver consistent, measurable results.
For an industry investing enormous sums in artificial intelligence, that may be the question that matters most.
Sources
- PYMNTS — Mercor Raises $750 Million for AI Training Platform — November 10, 2026.
- Mercor Official Blog — Organizing Human Intelligence for the Entire Economy — November 10, 2026.
- Reuters — AI Training Startup Mercor Valued at $20 Billion in Latest Funding Round — November 10, 2026.
- Business Wire — Mercor Raises $750 Million Series D at $20 Billion Valuation — November 10, 2026.
