Skip to content
@studio-11-co

falsify

PRML — Pre-Registered ML Manifest Specification · falsify.dev

falsify

PRML — Pre-Registered ML Manifests. An open, content-addressed format for committing a machine-learning evaluation claim to a SHA-256 hash before the experiment runs. Lock the threshold before the data, or it didn't happen.

A verifier recomputes the digest from the manifest's canonical bytes and emits a deterministic verdict: PASS / FAIL / TAMPERED. Moving the bar after the results exist produces a detectable hash mismatch. It proves the bar was locked, never the result.

Proof, not promises

4 reference implementations (Python, JavaScript, Go, Rust), byte-equivalent across 21 conformance vectors and rejecting a shared 14-vector negative suite. MIT code, CC BY 4.0 spec, ~600 tests, OpenSSF Best Practices passing.

Repositories

falsify Spec + Python reference + conformance suite (Go/JS/Rust under impl/)
falsify-js JavaScript reference (npm: falsify-js)
prml-verify-action GitHub Action: gate CI on tampered or regressed eval claims
mlflow-falsify MLflow plugin: bind runs to their locked bar, verify on run end
falsify-inspect Inspect AI adapter (on the UK AISI extensions page)
falsify-giskard Giskard scenario adapter
falsify-cookbook Patterns, anti-patterns, worked examples (CC0)
falsify-integrity-index Scorecard: 27 well-known eval claims vs 9 falsifiability criteria

Running a benchmark or leaderboard? Require a receipt — no locked claim, no entry.

Maintained by Cüneyt Öztürk · hello@falsify.dev

Popular repositories Loading

  1. falsify falsify Public

    A single-file Python CLI that pre-registers AI/ML accuracy claims with SHA-256. Lock the threshold before the data, or it didn't happen.

    Python 7 3

  2. prml-verify-action prml-verify-action Public

    Verify PRML (Pre-Registered ML Manifest) commitments in CI. Block merges on tampered or regressed eval claims.

    Python

  3. falsify-integrity-index falsify-integrity-index Public

    Public scorecard of how 27 ML eval claims meet 9 PRML falsifiability criteria. CC0 data; MIT tooling.

  4. falsify-inspect falsify-inspect Public

    PRML pre-registration adapter for Inspect AI eval logs. Listed on the UK AI Safety Institute's Inspect extensions page. MIT.

    Python 3

  5. falsify-cookbook falsify-cookbook Public

    Field manual for the PRML v0.1 specification. 13 patterns, 4 anti-patterns, 4 working examples. CC0.

    Python 1

  6. falsify-js falsify-js Public

    JavaScript reference implementation of the PRML (Pre-Registered ML Manifest) v0.1 specification. Byte-equivalent to the Python reference across 21 conformance vectors. Zero runtime dependencies.

    JavaScript

Repositories

Showing 9 of 9 repositories

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Loading…

Most used topics

Loading…