San Francisco startup Mercor is hiring skilled workers to train AI systems on professional tasks, betting $10 billion on the automation of white-collar jobs. Founded by college dropouts, the company represents a growing wave of AI firms targeting knowledge work.
Mercor's business model centers on a counterintuitive premise: employ high-skilled professionals to teach artificial intelligence how to perform their own roles. The approach mirrors how companies train AI systems across industries, but targets white-collar positions typically considered safer from automation.
The startup's founders, who dropped out of college, have attracted significant backing in a competitive AI funding landscape. Their $10 billion valuation places Mercor among the highest-funded AI companies, reflecting investor confidence in workplace automation technology.
The strategy involves detailed instruction on professional workflows—from software engineering to consulting tasks—creating datasets that improve AI model performance. Workers document processes, decision-making frameworks, and problem-solving approaches, which become training material for AI systems designed to replicate these capabilities.
Mercor operates amid broader industry transformation. Generative AI tools have expanded beyond creative and analytical work into roles requiring specialized expertise. The startup's focus on white-collar automation suggests a shift in AI's economic impact, moving beyond manufacturing and routine services into knowledge-based sectors.
Parallel developments underscore momentum in the sector. Thomas Reardon, who previously led Meta's Neural Band project, is raising funds for Flourish, an AI startup targeting a $2.5 billion valuation. Flourish focuses on developing energy-efficient AI systems, addressing computational costs that constrain current large language models.
The convergence of these efforts signals investor appetite for AI companies addressing deployment challenges and scaling automation. Whether Mercor's model creates economically viable AI systems or primarily generates training data remains unclear, but the capital influx indicates serious commercial interest.
These developments raise questions about workforce displacement and the timeline for widespread AI adoption in professional sectors. Industry observers point to ongoing technical challenges—particularly in reliability and domain-specific accuracy—that may slow automation deployment in critical roles.
Researchers have developed a typeface that remains readable to humans while evading optical character recognition systems. The font exploits weaknesses in how machine learning models process visual text.
Cosmetic surgeons report a growing number of patients requesting surgery based on AI-generated images, often with unrealistic expectations about achievable results.
US software development job postings on Indeed jumped 15% since Claude Code launched in February 2025, defying a broader 7% decline in overall job listings, according to Indeed Hiring Lab data.
Beijing Academy of Artificial Intelligence released Orca, a world model trained on 125,000 hours of unlabeled video that matches specialized robotics systems without ever seeing a single action label.