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, marking a stark contrast to the five months it took three highly skilled experts to do the same work manually. This efficiency gain sits at the intersection of surging commercial investments and a growing industrial demand for error-free software. For decades, developers have envisioned using math to prove that their software is bug-free, though this process remains long and laborious. Companies routinely run sophisticated tests to catch flaws, but formal verification—a mathematical approach capable of catching erratic behavior—remains rarely deployed due to high costs and operational friction.
The fundamental thesis driving the current market is that AI can automate the heavy lifting of formal verification, transforming it from an academic luxury into an essential corporate defense. At stake is whether organizations can escape the hidden bugs generated by rapid AI development cycles without inflating their engineering budgets, answering whether mathematical proofs can reliably replace traditional trial-and-error testing in commercial software pipelines.
The Commercial Push and Market Valuation
A wave of startups is currently capitalizing on this technological shift by raising massive capital rounds to bring automated mathematical proofs to mainstream programming. Among them is &Truth, which Stanford researcher Chuyue Sun joined in August and which publicly launched on Thursday. The startup has raised $70 million from investors and aims to eventually verify any programming language. Its research demonstrated the technology’s potential by using AI to verify part of the code that encrypts conversations in the messaging app Signal. Ilya Sergey, &Truth’s chief science officer, noted the broader industrial missing link by stating that bridges and buildings do not collapse because people did math before constructing them, asking why the industry does not do the same for code.
Bridges, buildings, they don’t collapse, because people did math before constructing them. Why don’t we do the same for code? The technology to mathematically reason was all there, just difficult to scale.
This argument is echoed by rival venture-backed players scaling the market. Rival Harmonic, which was cofounded by Robinhood CEO Vlad Tenev and has received $295 million in funding, announced this week that its math-focused AI agent Aristotle supports verifying software written in C and hardware designs in Verilog.
Enterprise Integration Realities
While high-profile tech infrastructure already relies on these methods—with Boeing planes, Amazon servers, and more than 2.5 billion active Apple devices containing at least some formally verified software—broad commercial adoption faces severe headwinds. Byron Cook, a distinguished scientist and vice president at Amazon Web Services, says formal verification has eased customer concerns about cloud security and helped internal teams move faster by squashing big bugs early on. Earlier this year, Apple’s security team said that some of the issues formal verification uncovered in its corecrypto library, which is used for encryption, could not have been found any other way. Yet, several midsized tech companies tell WIRED they have typically verified only one key feature of their products, if any at all, citing a lack of verification engineers, prohibitive costs, and fears that waiting weeks for code verification will stall product innovation.
Skeptics argue that the underlying mathematics remains fundamentally constrained by human and hardware unpredictability. Formal verification relies on making assumptions about how computers and people interact with software, leading some experts to question its ability to catch bugs stemming from user error and hardware breakdowns. As Vitalik Buterin put it in a blog post earlier this year, formal verification is not a panacea, even if it remains a necessary defense against automated cyberattacks. Whether this approach can escape being a trap that fails to replace exhaustive testing—and whether customers will pay for it—remains an open question for engineering budgets.
Review internal codebases this week to isolate high-risk encryption and security libraries where manual audits currently bottleneck release schedules, and evaluate whether pilot access to newly released verification add-ons can be integrated into existing staging pipelines.
Side by side
Where does the real boundary between costly mathematical certainty and mass software development lie? It sits squarely in compute budgets: startups like &Truth with $70 million in funding and Harmonic promise automated checks for hidden flaws, yet formal proofs have long remained a bespoke luxury even for tech giants. Amazon Web Services and Apple historically verified only critical slices like encryption, while mid-sized firms balked at scarce engineers and weeks of waiting. Mathematical proof of code is getting radically cheaper, yet enterprise pipelines are still built around testing a single critical feature and trusting trial and error for the rest.