Aefinity AI
Home
4 to 6 weeks

Paid pilots: do your receipts replay bit for bit elsewhere?

One of your inference pipelines emits CIS-2 receipts. We replay them on a machine separate from yours and report every mismatch. 4 to 6 weeks.

Email us about a pilot Start with the self-check

01Who it is for

Teams who need a check, not a promise.

Inference providers and platform teams

Who want a third party to be able to check that a deployment serves the model it says it serves.

AI governance and audit platforms

That want to attach checkable evidence of the exact computation to a record.

Evaluation and assurance teams

Who need results that others can replay.

02What happens

Four steps, one question.

The specification and verifiers are open (Apache-2.0); how much integration help you need is settled in scoping.

  1. 01

    Scoping

    We agree one pipeline, one open-weight model, a set of requests, and what a match must mean for you.

  2. 02

    Receipts

    Your pipeline emits CIS-2 receipts for those requests. The specification and verifiers are open (Apache-2.0); how much integration help you need is settled in scoping.

  3. 03

    Replay

    We replay the receipts on a machine separate from yours with the open verifier and compare every digest.

  4. 04

    Report

    You get a mismatch report that lists every request whose digests differ and, where the per-step trace is available, the earliest step at which it diverges; a replayable artifact (receipts, pinned digests, a verify script) your own staff can re-run; and a plain statement of limits.

03What you get

A written answer, and nothing to take on trust.

  • A written answer to one question

    Do your receipts replay bit for bit elsewhere, and if not, where do they stop matching?

  • Nothing you have to trust us for

    The verifier is open, and the artifact re-runs without us.

04What it is not

The limits, up front.

  • Not a correctness, safety or fairness audit.

    A matching receipt shows a computation was reproduced, not that the answer was right.

  • Not a certification, and not legal advice.

    We think the evidence is relevant to record-keeping requirements such as the EU AI Act's logging provisions; no auditor or lawyer has confirmed that mapping.

  • Not a promise that your stack will match.

    Our evidence is small open models in fp32, mostly on CPU, limited to greedy decoding. Production bf16 or GPU serving is outside the spec today, so the pilot shows how far your stack gets, and the report says so.

05Before you write

Run the self-check on your own machine.

See whether it matches. It takes about 15 minutes and needs no account or GPU.

selfcheck.shabout 15 min 
git clone https://github.com/Aefinity-AI/cis2-spec && cd cis2-spec && ./selfcheck.sh

The README lists what has been run, on which machines, and what has not. No unaffiliated party has reproduced it yet; every run listed is ours.

06How to start

Three lines is enough.

Paid pilots; contact us for scope and terms. Email justin@aefinityai.com with three lines.

AEFINITY AI INC. is an early-stage company in Orange, Texas, founded by Justin Thompson. Phone 409-320-1536.