AI-assisted formal verification draws startups and investment
AI systems have sharply reduced the reported time and cost of formal software verification, while startups including &Truth, Harmonic, Theorem and Axiom Math are expanding the field. &Truth launched publicly with $70 million in funding, and Anthropic and Theorem announced an AI-assisted project that found and patched hundreds of defects in 123 open-source libraries. Stanford researcher Chuyue Sun reported that her AI system completed the remaining specifications and proofs for an encryption library in 11.4 hours at a cost of about $467 after receiving the top-level specification. The reports differ on the human effort involved: WIRED described five months of work by three experts, while MIT Technology Review cited eight months by five major contributors. Axiom Math separately said AI agents produced 43.1 million lines of Lean-checkable formalized code for a mathematical proof in 24 days; that vendor claim was not independently validated. Formal verification remains experimental, and its guarantees depend on the correctness of specifications. Pricing, adoption and its ability to address user errors or hardware failures remain uncertain.
The reports describe AI substantially reducing the time and cost of a formal-verification project, alongside substantial financing, a public launch and newly announced tools. These developments could make formal methods more accessible for critical software, but broad adoption and the limits of automated verification remain unresolved.
Stanford researcher Chuyue Sun and her fellow computer scientists developed an AI tool that completed an engineering project in less than half a day for under $500, compared to the five months it took three highly skilled experts to do the same work manually. The dramatic reduction in time and cost, which Sun reported in a recent paper, is among the early results that are awakening an obscure field of coding and mathematics known as formal verification.
For decades, developers have envisioned using math to prove that their software is bug-free. This process is long and laborious, requiring engineers to write precise statements about the goals of their software and convert their code into formulas that can be validated. While companies already run sophisticated tests to identify as many flaws in their software as possible, they rarely shell out for formal verification, which can catch even more erratic behavior.
Now, a wave of startups is betting that generative AI will make formal software verification both practical and essential. Among them is &Truth, which Sun joined in August and which publicly launched on Thursday. Sun’s research demonstrated the technology’s potential by using AI to verify part of the code that encrypts conversations in the messaging app Signal.
#&Truth#Harmonic#Axiom Math#Theorem#Anthropic#Claude#Aristotle#Chuyue Sun#Lean#CodeRabbit#Signal#Pramaana Labs