Pramaana Labs raises $27M seed round from Khosla Ventures to bring formal verification to AI

In an era where businesses are striving to transform AI pilot programs into integral components of their operations, the issue of reliability has emerged as a top priority. Enter Pramaana Labs, a startup determined to tackle this challenge by leveraging the power of mathematical formalization—a robust system from computer science—to bring order to the often chaotic world of artificial intelligence. On Wednesday, Pramaana Labs revealed that it has successfully raised $27 million in seed funding, spearheaded by Khosla Ventures, alongside contributions from prominent firms including Accel, Boldcap, Nexus Venture Partners, Premji Invest, and Unbound. The startup is set to target critical sectors such as law, drug discovery, and tax preparation, where the cost of errors can be significant and the demand for reliability is paramount. As AI technology integrates into these sensitive areas, there is an urgent need for enhanced safeguards against inaccuracies and misinterpretations. Ranjan Rajagopalan, co-founder and CEO of Pramaana, emphasizes that these fields are particularly suited for formalization. He likens the tax code to mathematics, explaining, "It’s like math in the sense that you have a lot of rules that you need to abide by. Once you have a codified version of it, the reasoning on top of it starts becoming deterministic." Pramaana's innovative system operates on a conventional large language model (LLM), which allows it to respond to natural language inquiries and tackle complex challenges that traditional computing methods may struggle with. However, what sets Pramaana apart is the incorporation of a deterministic layer atop the LLM, ensuring the accuracy of its outputs. The fusion of LLM technology with formal verification is gaining traction, but Pramaana's distinctive strategy involves utilizing the formal verification tools found in the open-source LEAN programming language, which is known for verifying mathematical proofs. Rajagopalan cites the CATALA project in France as a successful example of formalizing the nation's tax and benefits system into executable code. For each application, Pramaana plans to develop tailored LEAN-style formal verification systems, guided by experts in the relevant fields. In its pursuit of excellence in tax law, the company collaborates with former IRS commissioner Danny Werfel, while distinguished professors from IIT Delhi, IIT Madras, and UC Berkeley are advising on cybersecurity and drug discovery applications. Rajagopalan concludes with a powerful statement: "The world’s hardest problems are not unsolvable. They are unformalized. Every domain where being wrong can cost someone their health, money, or freedom has rules. Now, those rules just need to be codified."