AI Formal Verification Company Pramaana Labs Raises $27 Million in Seed Funding, Introducing LEAN Mathematical Proofs to Address Hallucinations

According to Beating Insight Monitoring, the AI startup Pramaana Labs, dedicated to bringing formal mathematical verification into AI, announced the completion of a $27 million seed round. This round was led by Khosla Ventures, with participation from Accel, BoldCap, Nexus Venture Partners, Premji Invest, and Unbound. The funds will be used to build a deterministic verification layer for highly sensitive, error-intolerant industries such as legal, pharmaceutical R&D, and tax reporting.

At the core, Pramaana Labs’ underlying system still runs traditional LLMs to maintain flexibility in handling complex problems and natural language, but its key innovation is that it builds a deterministic verification mechanism on top of the LLM. The mechanism takes inspiration from the open-source LEAN programming language used to verify mathematical theorems, and codes industry rules for specific domains.

To develop LEAN-style formal verification systems for different vertical industries, Pramaana has invited a large number of domain experts to oversee and endorse the work. In the tax sector, the company is collaborating with former U.S. Internal Revenue Service (IRS) commissioner Danny Werfel; in cybersecurity and pharmaceutical R&D systems, it is led by a team of professors from the Indian Institutes of Technology (IIT) Delhi and Madras, as well as the University of California, Berkeley.