Auditable citation integrity, at publication scale.

Vahtian builds the sentinel layer for research citations: a blinded human → AI → adjudication workflow with a decision-gated, undoable write and a hash-chained audit trail. Free and local-first for researchers; paid infrastructure for the organizations that carry citation-risk exposure.

vahti — Finnish for sentinel / guard. The watch over what gets cited.

The Vahtian products

CiteVahti Researchers & labs — verify the claim before you cite it, inline in your manuscript. live

ReviewVahti Journals & editorial offices — claim-level checks, reviewer oversight, manuscript audit trails. roadmap

GuidelineVahti Guideline groups, societies, institutions & medical-communications teams. roadmap

AtlasVahti Consented, de-identified evidence-validation patterns — governed analytics & API. roadmap

What every Vahtian product holds to

The human decides

AI is a blinded, advisory second rater — never decisive, never silently propagated. Disagreements route to a person.

No silent writes

References enter your library only as the audited, undoable last step of a human decision. Dedupe fails closed.

The reasoning is recorded

A hash-chained audit trail, so the workflow can be reported transparently in a methods section.

Open core, local-first

The core is Apache-2.0 and runs on your laptop. We never paywall a researcher’s ability to verify their own work.

Contact

General & partnerships: hello@vahtian.com
Code: github.com/heidihelena/citevahti