Midas, an AI security and reliability company focused on mathematical verification, has launched publicly after closing a $10 million funding round led by Valor Equity Partners and Nova Global. The company says it is building a “mathematical trust infrastructure” to ensure AI outputs and underlying reasoning are provably correct before they are used in real-world systems where mistakes are costly.
The company is positioning its platform to address a central AI adoption gap: models can generate fluent responses, but they do not inherently provide proof that those responses are correct. Midas says it applies formal verification to AI outputs, data, and reasoning, with the goal of embedding evidence and correctness checks into AI workflows rather than relying on after-the-fact review.
Midas says it was founded by a team with deep competitive mathematics and computing backgrounds, including medalists from the International Mathematics Olympiad and the International Olympiad in Informatics, as well as engineers with experience at firms including Jane Street, Google, AWS, NVIDIA, and Mercor. The company also notes academic backgrounds spanning Stanford, MIT, Cambridge, Princeton, and Duke.
With the new capital, Midas plans to translate formal verification research into production-grade infrastructure and pursue deployments across mission-critical sectors. The company cited target environments including biotech, defense, hardware design, financial systems, and underlying AI and cloud infrastructure, where provable correctness is required rather than optional.
Investors participating in the round framed verification as a missing layer in the AI stack, arguing that as AI systems scale and outputs outpace human evaluation, infrastructure to enforce correctness will be necessary for safe deployment. Midas is presenting its approach as proof-native, emphasizing verification at the start of an AI process rather than error-checking at the end.
KEY QUOTES
“Modern AI produces fluent, convincing answers, but it cannot prove they are correct. Midas is building the barrier between probabilistic outputs and real-world systems. We enforce correctness mathematically, so results are not inferred, argued, or hoped for, but proven before they are allowed through.”
Shalim Monteagudo-Contreras, President and Co-Founder, Midas
“Every human institution, from law to science to finance, runs on evidence. Artificial intelligence is the first form of intelligence that operates without it.”
Renzo Balcazar, CEO and Co-Founder, Midas
“Every layer of the AI ecosystem has eventually required its own infrastructure. Verification is the final missing layer. This is not about probabilities, but proof. What sets Midas apart is its culture: a team trained to reject ‘almost correct’ answers and accept only what can be demonstrated.”
John Stanton, Vice President, Valor Equity Partners
“At Nova Global, we focus on backing founders with the potential to become historical figures. Shalim Monteagudo-Contreras and Renzo Balcazar are already operating at that level. They’ve built a world-class team from scratch and are taking on one of the most fundamental challenges in AI: trust. Their proof-native approach to ensuring AI reliability demonstrates both the technical ambition and founder-market fit is what turns Midas into a generational company.”
Carlo Agostinelli, Founder, Nova Global
“According to Rodrigo Porto, Tech Lead at Midas, verifying reasoning from the start, rather than checking errors at the end, is what makes trust possible as systems grow too complex for manual review. Midas introduces mathematical evidence at the core of AI, verifying outputs, data, and reasoning so these systems can be trusted where mistakes are not an option.”
Rodrigo Porto, Tech Lead, Midas
