Who wrote this?Every word. Confidence-rated.
Three modes of authorship analysis. A copyright similarity check. An interactive process map. And a confidence rating on every result, because honest tools don't pretend certainty they don't have. It rates how sure the analysis is about itself, never how honest you are.
Three ways to articulate authorship
Each mode returns a confidence rating on a spectrum from Highly Conflicted Analysis (pale pink) through Moderately Confident Analysis (yellow) to Highly Confident Analysis (green). The limits of the analysis come with the findings. You deserve to know both.
Full phase-by-phase slider control, tool declaration and submission context. Best for course submissions and institutional records.
Five inputs. The engine infers phase involvement from your description and sample. Best for journal submissions and grant applications.
Stylistic and linguistic similarity analysis against known copyrighted work traditions. Every finding confidence-rated.
Detailed Phase Control
You set AI involvement per phase. The engine applies genre-weighting and returns a confidence-rated analysis. It's that direct.
Phase-by-phase sliders
Set AI involvement across all 8 phases from 0 (none) to 10 (fully AI).
Tool declaration
Specify which AI tools you used, for your own record and for institutional transparency.
Confidence rating
Green, yellow or pale pink: each signals how certain the analysis is, with a detailed breakdown of what's driving it.
Process map upgrade
A colour-coded flow of human vs. AI interaction across all phases. Optional but worth enabling.
Process Map Upgrade
Interactive colour-coded map of human-AI interaction across all 8 phases.
Independent Signal Check (optional)
A dry run of how a naive AI-detector might read your text, so you're never blindsided by one. This isn't a verdict on your work; automated detectors are famously unreliable and misread honest writers (especially multilingual authors) all the time. Think of it as a weather report, not a diagnosis. If it flags something, that's a prompt to add context to your disclosure, not a mark against you.
Five inputs. Full intelligence.
Phase scores are inferred automatically. The confidence rating reflects how much signal your description provides. Be specific and it'll reward you.
Inferred phase scoring
The engine reads your process description and assigns phase scores. The more you give it, the more it can work with.
Voice analysis
Your sample is assessed for stylistic consistency and authentic voice markers.
Confidence rating
Shorter or vaguer descriptions typically yield yellow or pale pink confidence. Longer, specific ones earn green. It's not about length; it's about specificity.
Process map upgrade
Inferred phases visualised as a colour-coded human-AI flow map.
Process Map Upgrade
Visualise inferred human-AI interaction as a colour-coded flow map.
Independent Signal Check (optional)
A dry run of how a naive AI-detector might read your text, so you're never blindsided by one. This isn't a verdict on your work; automated detectors are famously unreliable and misread honest writers (especially multilingual authors) all the time. Think of it as a weather report, not a diagnosis. If it flags something, that's a prompt to add context to your disclosure, not a mark against you.
Copyright Similarity Check
Full copyright analysis with confidence rating is in the dedicated section below. It'll only take a moment.
Go to Copyright Check ↓How original is your work?
Stylistic, structural and linguistic similarity analysis, with a confidence rating on every finding. This is a literary analysis aid, not a legal determination.
Stylistic similarity
Sentence structure, lexical choices and narrative rhythm that echo known works.
Genre-specific signals
Risk is calibrated for your genre; poetry signals differ from long-form fiction, and the engine knows that.
Confidence-rated findings
Each similarity signal rated green, yellow or pale pink by the engine's certainty about what it's found.
Originality recommendations
Four targeted suggestions. Developmental, not punitive.
This is a literary analysis aid, not legal advice. Always consult qualified legal counsel for anything that matters legally.
This tool uses water too.
The RBA runs on the same kind of AI it helps you talk about, and that has a real water and energy cost. We show you ours because a transparency tool that hid its own footprint wouldn't be much of one. No lecture. Just the number.
Find nonpartisan literary and conservation organisations near you, searchable by ZIP code. No agenda, just a good directory.
Open directory →What we do and don't do with your work samples. No ambiguity, no boilerplate. Just a straight answer.
Read the notice →