DkMath — Verifiable AI Mathematical Research

The description states that DkMath is:

• A large formal mathematics research library built on Lean 4 and Mathlib.
• A system that uses AI tools (GPT-5.6, Codex) to assist in theorem discovery and implementation.
• A workflow that includes Git for provenance, GitHub Actions for CI, Manim and FFmpeg for visualization, and Lean 4 for verification.

It is not clear whether DkMath is a standalone product or an experimental research tool. The author describes it as a “formal mathematics research library” and a “workflow,” but does not define a distinct software offering or service.

Inference The project appears to be a proof-of-concept demonstration of how AI can be integrated into formal mathematical research, rather than a commercial product.

The description states:

• DkMath is positioned as an “auditable human–GPT-5.6–Codex workflow.”
• It treats AI not as an authority but as part of a disciplined research loop.
• The system verifies theorems using Lean 4 and Mathlib, and visualizes results with Manim.

Inference The positioning is that DkMath is a tool for formal mathematical research that integrates AI in a controlled, verifiable way. It does not appear to be positioned as a general-purpose AI assistant or marketplace.

The description does not identify any specific customer base or target market. The author refers to the project as a “research library” and a “workflow,” but does not name users or customers.

Inference It is unclear who the intended users are — whether they are mathematicians, formal verification researchers, or academic institutions. No evidence of customer segmentation or ICP is provided.

There is no evidence in the description of a business model or pricing structure. The project is described as a research tool and not as a commercial offering.

Inference No business model or pricing information is evident from the self-reported description.

The description states:

• DkMath uses Lean 4, Mathlib, Git, GitHub Actions, GPT-5.6, Codex, Manim, FFmpeg.
• The system includes a pipeline that integrates human direction, AI review, code generation, verification, and visualization.
• It demonstrates a theorem chain involving finite primes and GN theorems.

Inference The technical stack is well-defined for formal mathematics and AI-assisted research. However, no evidence of scalability or production-grade delivery is provided.

The description states:

• A public demo was created and published on YouTube.
• The project includes a release (m-v1.0.0).
• It demonstrates a theorem chain and visualizes it with Manim.
• The system passed Lean CI and was merged into the main branch.

Inference There is evidence of a working prototype, but no evidence of adoption, usage metrics, or commercial traction beyond the author’s own demonstration.

The description does not mention any competitors. It does not state whether similar tools exist in the market for formal mathematics or AI-assisted theorem proving.

Inference No competitive landscape is described. The project appears to be self-contained and not positioned within a known marketplace.

• The description is entirely self-reported, with no independent verification.
• No evidence of revenue, customers, or adoption.
• The project is described as a research tool, not a product.
• The use of GPT-5.6 and Codex implies reliance on proprietary AI models that may not be available to others.
• No indication of scalability or production deployment.

Inference The lack of commercial traction, customer data, or market positioning raises questions about the project’s viability as a business or product.

1. What is the intended use case for DkMath beyond this demonstration?
2. Are there any plans to monetize or commercialize the system?
3. How does the system scale beyond this single demonstration?
4. Is there any evidence of adoption by academic or research institutions?
5. What are the limitations of using GPT-5.6 and Codex in formal verification workflows?