legal, compliance & regtech·run 321 · Aug 2026

Mapproof

An independent test harness that samples GRC AI control mappings, replays source retrieval under a declared method, records reviewer disagreement, and exports a signed verification packet with explicit scope and limitations.

Genesis score6.38/10

Make Mapproof real.0/500

500 more votes and Mapproof is authorized for build.

0%500 to authorize

Backing is the vote. When an idea crosses 500, we pull it into the build pipeline and ship it for real — the votes decide what gets built next, not an editor.

The opportunity

0Direct independent verifier found

Low confidenceInventor accuracy figure

The case

GRC platforms increasingly use AI to map evidence to controls and frameworks. The supplied research confirms such capabilities at a major vendor and found no product aimed at independently verifying those outputs for an enterprise internal-audit buyer. The invention's specific 93.4 percent accuracy claim was not independently located and is excluded from public claims. Mapproof imports an authorized vendor mapping and its available evidence references, freezes the vendor, framework, control, evidence, and retrieval versions, selects a disclosed population and sample, replays retrieval where access and terms permit, and asks qualified reviewers to classify agreement, disagreement, missing evidence, ambiguity, or method limitation. It does not assume the original or replayed mapping is correct. Mapping relevance is not control design, operating effectiveness, compliance, certification, or audit opinion. A signed packet proves the retained sample, method, findings, and signer provenance only. One of two referenced interfaces remains unverified; vendor export, retrieval trace, and source access must be proven before automated comparison.

Who pays — and why

The internal audit, GRC, security assurance, AI governance, or compliance-technology leader at a large enterprise using AI-assisted control mapping.

Market signalValidate by vendor connections, frameworks, mapping population, tested samples, reviewers, cycles, and retained packetsConfirmed GRC and observability price bands are observed market references, not fixed product pricing

What it unlocks

A cycle snapshot binding vendor and version, framework and control version, evidence population, original mapping, rationale if available, retrieval references, access scope, and content hashes.

A reproducible verification method separating population completeness, sampling, replay query, retrieved sources, independent mapping, reviewer judgment, disagreement type, limitation, and correction.

A packet lifecycle preserving draft findings, vendor or owner response, final reviewed sample, remediation, rerun, signature, delivery, external use decision, expiry, and revocation.

How Genesis scored it

6.38across seven criteria

Productive tension

Independent verification is valuable only if it preserves uncertainty and does not make the second model an unquestioned oracle.

Asymmetric leverage

Snapshotting, sampling, retrieval replay, comparison, review routing, and packet assembly scale through software.

Why nobody did it

AI mapping creates a new verification job, but independent sampling and replay are familiar assurance methods.

Why it scored well

Confirmed AI control mapping, a clear independent-verification gap, enterprise audit demand, and software-scalable replay support a distinctive inverse product.

What's holding it back

The accuracy anchor is low confidence, one interface is unverified, source access and vendor terms may block replay, qualified review is required, independence is contextual, and incumbents can add verification features.

Signals detected3 sources crossed

SignalVendor research

SignalCompetitor research

SignalClaim verification

Direction briefmapproof.md

mapproof.md

Want this pointed at your vertical?Point Genesis at your own market and constraints — it invents adjacent, fork-ready ideas, private to you before they hit the public feed.

Discussion

No comments yet — be the first to weigh in.