Aptos says it has become the first layer‑1 to combine AI with formal verification for dynamically scheduled Move contracts — a claim that, if borne out, pushes smart‑contract safety into new territory as machine agents increasingly write and trade code in real time.
In a post on X, Aptos Labs laid out a workflow in which large language models draft formal specifications, and the Move Prover then mathematically verifies those specs against contract code. “AI writes the specifications, mathematics proves their correctness, and the Move Prover serves as the oracle — this is the trust layer between the market and the machine,” the team wrote, positioning the prover as a gatekeeper between automated agents and on‑chain markets.
Why this matters: Move has long touted native formal verification as a core strength, and Aptos says the toolchain now supports dynamic dispatch and first‑class functions — features that let contracts treat functions as values and call code determined at runtime. That flexibility dramatically expands the state space and verification complexity, so Aptos engineers redesigned how the Move Prover generates and checks verification conditions for higher‑order Move code. The architecture change is documented in a May 2026 paper titled “Formal Verification of Imperative First‑Class Functions in Move.”
Aptos’ Move Prover is already used at the protocol level to verify critical logic — staking, metering, code deployment and underlying data structures — and the team emphasizes usability for domain experts with mathematical backgrounds who may not be seasoned software engineers. The claim is that combining LLMs to produce formal specs with an automated prover can scale auditing and enforcement as contracts become more dynamic and machine‑driven.
The announcement comes alongside Aptos’ $50 million program to fund on‑chain markets and AI systems. The company has also promised privacy‑focused features such as encrypted mempools and confidential perpetuals, pitching “institution‑friendly” trading rails designed to withstand adversarial bots and next‑gen automated strategies. In that context, framing Move Prover as an “oracle” is both a technical claim and a strategic argument: formal methods, Aptos contends, are the scalable way to police smart contracts once AI agents are routinely authoring, deploying and trading against them.
Read more AI-generated news on: undefined/news
