Quantum Computing
Zapata and UMD Launch ForShor for Verified AI-Assisted Quantum Research

Key PointsForShor requires AI-proposed circuit designs to pass formal Lean proofs before earning…
View more
- ForShor requires AI-proposed circuit designs to pass formal Lean proofs before earning leaderboard credit.
- The challenge targets circuit optimization for Shor's algorithm rather than proving hardware can factor large numbers.
- Zapata will use ForShor as a proving ground for verification methods in its Quantum Pilot platform.
Zapata Quantum and the University of Maryland have introduced ForShor.ai, an open challenge for AI-assisted quantum research built around a formally verified implementation of Shor’s algorithm. Participants can use AI to propose improvements, submit solutions to a leaderboard, and explore documentation linking the mathematics to machine-checked proofs.
The initiative targets a growing problem in AI-driven science: producing a plausible algorithm or explanation is easier than establishing that it is correct. ForShor places the search for better designs inside a framework with explicit mathematical constraints, while keeping researchers involved in understanding the results.
Announced November 18, the project extends Zapata’s collaboration with UMD and is intended to test verification techniques relevant to the company’s Quantum Pilot software platform.
Why Shor’s Algorithm Is the Starting Point
Shor’s algorithm is best known for its implications for cryptography. Its quantum component can find a repeating structure, called an order, in modular arithmetic. Classical mathematical steps can then use that result to help factor a large integer.
ForShor focuses on the design of the arithmetic circuits involved in that computation. The University of Maryland’s project description identifies fast Toom–Cook multiplication as a key building block. This approach breaks a large multiplication problem into smaller pieces, creating choices about how to organize the arithmetic.
Those choices affect how many gates and working qubits a circuit needs. A gate is an operation in a quantum circuit; a logical qubit is a unit of quantum information at the algorithmic level. Reducing either requirement can improve a design’s prospects, although translating it into physical hardware introduces further costs.
The challenge gives researchers a specific optimization problem with checkable conditions, rather than asking an AI system to invent an entirely new quantum algorithm without constraints.
What the Verification Framework Checks
The ForShor preprint, submitted November 16 by Anirudh Suresh, Joshua M. Courtney, Runzhou Tao, and Yudong Cao, describes a harness that accepts a design specification called a policy. That policy defines choices in a recursive circuit construction and can come from a human researcher or an AI agent.
The framework covers three connected claims: the implementation’s mathematical correctness, the asymptotic growth of its gate cost, and the correctness of translating the verified algorithm into a more detailed circuit representation.
The reported gate-cost scaling is O(n2+ε), where n describes input size and ε is positive. This expresses how the cost grows as the problem gets larger. It is not a statement that a particular 2,048-bit instance can be run with a small fixed number of gates.
The implementation uses Lean 4 formal proofs. In formal verification, researchers state the required behavior precisely and supply proofs that a proof assistant can check. This goes beyond testing a program on selected examples, although the guarantees remain tied to the definitions and assumptions being proved.
AI Proposes Designs, the Framework Checks Them
The project’s interactive proof documentation shows how arithmetic choices fit into the larger order-finding circuit. It connects explanations and diagrams to the corresponding Lean code so readers can inspect the argument at different levels of detail.
One part of the design space concerns instruction tables for recursive multiplication. Different arrangements of additions, shifts, and related operations can change a circuit’s cost. A proposed table must satisfy the framework’s admissibility conditions before it can inherit the relevant correctness guarantees.
This division of work is useful for AI-assisted research. An agent can explore candidate arrangements or suggest optimizations, while an independently checkable contract determines whether the result is valid. Researchers can then study why a successful candidate works, rather than treating its score as the only output.
“ForShor puts formal verification, with humans in the loop, at the center of that process,” said Peter Shor, the algorithm’s creator and a member of Zapata’s Scientific Advisory Board.
Zapata CTO Yudong Cao described the initiative as a white-box learning and research environment. The public challenge includes a leaderboard and community resources, and the company says contributors may become co-authors of a subsequent research paper.
How ForShor Connects to Quantum Pilot
ForShor is a focused research challenge within a broader effort to make quantum application development more systematic. Zapata’s Quantum Pilot platform combines a knowledge layer, computational tools, and verification workflows.
Quantum Graph connects applications, algorithms, research, and hardware roadmaps. Quantum Engine handles work such as algorithm selection, resource estimation, experiment design, and benchmarking. Quantum Assurance combines formal methods with expert review before results inform a decision.
Resource estimation asks what an algorithm would actually require to execute. A mathematically valid design can still demand more qubits or runtime than an available machine can provide. Keeping that analysis connected to the verified implementation helps researchers avoid proving one circuit correct while estimating the cost of a different one.
As covered in Unite.ai’s earlier Quantum Pilot article, the platform is intended to connect potential use cases with technical feasibility and comparisons against classical computing. Zapata says ForShor will serve as a proving ground for verification methods used in that wider process.
A Proof of Software Properties, Not Hardware Readiness
ForShor’s contribution is an inspectable software and mathematics framework. It does not demonstrate that today’s quantum processors can factor cryptographically significant numbers or eliminate the engineering demands of fault-tolerant computing.
Physical error rates, error correction, hardware connectivity, and execution time remain separate considerations. Likewise, a machine-checked proof establishes a stated mathematical result; making that result understandable to researchers still requires clear definitions, documentation, and review.
The broader research idea is to let AI explore designs aggressively while requiring its proposals to pass explicit, reproducible checks. ForShor gives that approach a concrete starting point: a defined quantum task, an open implementation, and a community that can inspect both the candidate designs and the reasoning behind them.












