Harmonic Raises $100 Million to Develop Mathematical Superintelligence

California-based startup Harmonic announced on July 9, 2025, that it had raised $100 million in Series B funding, bringing its valuation to nearly $900 million. The company, which focuses on developing mathematical superintelligence, aims to solve complex mathematical problems that current AI cannot handle.

Record Investment in Specialized AI

The funding round was led by Kleiner Perkins, with significant participation from Paradigm. Other investors included Ribbit Capital and existing backers, including Sequoia Capital, Index Ventures, and Charlie Cheever. Its valuation of $875-900 million ranks Harmonic among the most valuable AI startups in the field of specialized artificial intelligence.

This funding follows a previous Series A round in September 2024, when the company raised $75 million in a round led by Sequoia Capital, with significant participation from Index Ventures and additional backing from DST Global partners and Nikesh Arora.

Mathematical Superintelligence vs. Traditional AI

Harmonic focuses on developing mathematical superintelligence (MSI), which is intended to surpass human capabilities in solving mathematical problems. Unlike language AI models that are prone to hallucinations, Harmonic's Aristotle system uses MSI to verify accuracy, ensure logical reasoning, and transparently identify errors and inconsistencies.

"Aristotle's MSI is uniquely suited for critical applications where there is no room for error, such as generating verified software and formally verifying existing code—a breakthrough for industries including blockchain, financial services, aerospace, and other safety-critical systems," explained Tudor Achim, CEO of Harmonic.

Leadership and Vision

Harmonic was founded in 2023 by Tudor Achim and Vlad Tenev and is headquartered in Palo Alto, California. Tudor Achim, who serves as CEO, previously co-founded and served as CTO of autonomous driving startup Helm.ai. Achim earned a bachelor's degree in computer science from Carnegie Mellon University and was a doctoral candidate at Stanford.

Vlad Tenev, co-founder and executive chairman of Harmonic, also serves as CEO and co-founder of Robinhood Markets. Tenev earned degrees in mathematics from Stanford (B.S.) and UCLA (M.S.).

"We are approaching AI that can truly reason—quickly, reliably, and at scale," said Vlad Tenev. "Mathematical superintelligence is moving from research into real-world applications, and we are excited for Aristotle to begin putting this AI capability into users' hands."

Aristotle: A Flagship Model with Practical Applications

The Aristotle model is designed to solve mathematical problems at a level far beyond human capabilities. According to the company's official statement, Aristotle is already helping researchers in theoretical mathematics, and the company believes that it will soon contribute to solving major open problems not only in mathematics but also in related fields such as physics and computer science.

Accelerating mathematics opens the door to breakthroughs in several dependent fields, including theoretical physics and engineering. The company plans to make its flagship AI model, Aristotle, available to researchers and the public later this year, with an emphasis on useful and enjoyable real-world applications.

Strategic Importance for the Future of AI

Ilya Fushman, a partner at Kleiner Perkins and former physicist who will join Harmonic's board as an observer, commented: "Harmonic has created a new foundation for verified, scalable reasoning that can be trusted in high-stakes environments. I am deeply excited about Aristotle's applications not only in software, but also in accelerating progress across science, engineering, and general intelligence."

He joins existing board member Andrew Reed of Sequoia and existing board observer Jan Hammer of Index.

The Future of Mathematical Reasoning and Team Expansion

Harmonic represents one of the first major attempts to create mathematical superintelligence, which it defines as AI capable of solving mathematical problems at a level beyond any human. The company's broader goal is to push the boundaries of human understanding by using AI for verified and interpretable reasoning in mathematics and related fields.

"The near-term goal is to build AI that solves mathematical problems at a level superior to any human," said Tenev. "The ultimate goal would be to solve major unsolved mathematical problems and extend that to problems in physics and computer science."

At the same time, the company is actively hiring new talent to build and commercialize mathematical superintelligence, indicating ambitious plans to rapidly scale its team and accelerate development.