About Genutext

    A practical review tool built around privacy and restraint

    Genutext gives individual educators access to AI detection and source matching without requiring a subscription or institutional contract.

    Why it exists

    Detection should support a decision—not become the decision

    AI-assisted writing has made authorship questions harder, while a single automated score can look more certain than it is.

    Genutext was built to make a focused checking workflow available to teachers, tutors and other reviewers who may not have access to an institution-wide product. Users can try a short AI-only preview, then buy individual scan credits when longer documents or source matching make the extra context worthwhile.

    The product is developed and operated by a UK-based sole trader. It combines an external analysis provider with a Genutext interface for score interpretation, sentence-level review, source candidates, temporary detailed results and optional email reports.

    Product principles

    Four constraints shape the service

    Privacy before convenience

    Submitted writing is processed for the scan and not kept in Genutext's server-side scan history. Detailed results remain temporary unless the user requests an email report.

    Pay only when needed

    Standard and extended credits are purchased per scan, do not expire and avoid a recurring subscription.

    Separate distinct questions

    AI-writing signals and plagiarism source matching are presented as separate results because authorship patterns and source overlap are not the same thing.

    Keep a person accountable

    Guidance throughout the product frames automated outputs as indicative and directs reviewers back to context, evidence and professional judgement.

    Transparency

    Document the boundaries as carefully as the features

    A useful detection product should explain what it analyses, how labels are formed and what the result cannot establish.

    The methodology page describes the current English-language analysis pipeline, AI-score conversion, paid sentence aggregation, source-match handling and storage boundary. It also states that Genutext does not currently publish a universal product-accuracy percentage.

    The responsible-use resource library covers false positives, score interpretation, evidence standards, source similarity and fair coursework review. These pages are intended to make the limitations visible before a result is used in a consequential decision.

    Questions or feedback

    Help us make the boundaries clearer

    Contact Genutext about the product, billing, account access, privacy or the responsible-use guidance.