The founder of one of the fastest rising neolabs is sharing her perspective on the war over AI talent.
Loading audio narration…
Carina Hong, founder of the startup Axiom Math, which focuses on building an AI mathematician, told Business Insider’s Ben Bergman that some researchers see advantages in working for neolabs rather than larger tech companies at Business Insider’s The Long Play on Tuesday.
“We had really amazing talents coming from Big Tech. They are industry veterans,” said Hong. “They are a bit tired of the PSC culture or all the politics at big shops, and they just want to work at a place where they feel like it’s not sort of like job security because it is a startup — there’s no job insecurity — but more like security for the research direction.” (PSC refers to the rigorous biannual performance review process at companies like Meta, where employees are evaluated by their peers and managers.)
“When you’re in a big lab and your research direction changes on like a monthly, if not weekly basis, that’s actually another sort of security they are pursuing,” Hong added. “You want to work at a place where you know that what you’re working on still has impact in like two months.”
AI-focused companies, from startups like OpenAI to Big Tech companies like Meta, are offering massive paydays to AI talent.
Axiom Math is developing advanced AI systems that can not only solve complex math problems but also generate step-by-step proofs and verify their own reasoning, Hong said. She told Bergman that the broader goal is to use mathematical reasoning as a foundation for more powerful AI, potentially enabling applications in areas like software verification, finance, and cryptography.
Axiom Math is only about a year old and is part of a new class of startups known as “neolabs,” gaining traction across the AI sector, particularly in hubs like San Francisco. These companies are typically small, fast-moving teams building AI-native products from day one, often with fewer employees, mixed-use office spaces, and significantly lower operating costs than traditional startups. As of March 2026, Axiom Math is valued at $1.6 billion.
