Know Nota

Know Nota is a writing analysis platform for people who need a cautious, evidence-based view of how a document was written. It examines how machine-like a text appears, traces the origins of its words, and can create a signed record of the writing process that others can verify.

The platform is designed for students, educators, schools, and organizations that need more context than a simple AI detection label. Each scan presents the score alongside its confidence limits, including how often that specific model produces incorrect results. Know Nota also avoids claiming that a document was definitively written by AI or identifying a particular AI generator.

Key Features

Evidence Alongside Detection Scores

Know Nota puts the limitations of its analysis directly alongside the result.

Each scan shows:

• Detection score
• Result band
• False-positive rate for the specific model

This provides additional context when interpreting a score rather than presenting the result as definitive proof of authorship.

Short-Text Abstention

Know Nota does not attempt to score very short passages.

For text under 100 words, it refuses to provide a detection score and explains why the available evidence is insufficient.

This approach is designed to avoid presenting an unreliable result when there is too little text to analyze.

Document-Level Analysis

The platform evaluates writing at the document level rather than producing sentence-by-sentence detection maps.

This keeps the analysis focused on the overall characteristics of the submitted text instead of treating individual sentences as independent evidence.

Source & Writing Records

Know Nota can create a signed record showing how a document was written.

The record can include a link that other people can use to check the evidence, making it useful when a student, educator, organization, or reviewer needs documentation alongside an assessment.

Labeled Evidence

Source checks are separated into labeled components so users can distinguish between different types of analysis.

The output identifies what was:

• Inferred
• Measured
• Matched
• Observed

This makes the results easier to interpret when they are being used in educational, editorial, hiring, or internal review settings.

Privacy by Default

Pasted text is discarded unless you explicitly choose to keep it.

This provides a privacy-focused workflow for users who need to analyze writing without automatically creating a permanent copy of every submitted passage.

Google Docs & File Review

Review writing from existing documents rather than relying only on pasted text.

Know Nota supports:

• Google Docs
• .docx files
• PDF files
• Text files

This makes it suitable for reviewing complete documents in the formats people already use.

Batch Checks & API Access

Teams that need repeated analysis can use batch processing and API access.

These options support larger-scale workflows while retaining the same caveats and contextual information attached to individual results.

Built For Students, Educators & Organizations

Know Nota is designed for situations where writing analysis needs to be accompanied by evidence and appropriate limitations.

It is particularly useful for:

• Students documenting or reviewing their writing process
• Educators assessing writing with additional evidence
• Schools developing consistent review workflows
• Editorial teams examining submitted writing
• Organizations conducting internal writing reviews
• Hiring teams reviewing written materials with appropriate context

Common Use Cases

• Academic writing review: Examine writing while keeping detection limitations visible.
• Student work: Provide evidence around how a document was written rather than relying only on a score.
• Educational review: Give educators additional context when evaluating submitted work.
• Editorial workflows: Review documents and their sources using labeled evidence.
• Hiring review: Add writing analysis to an internal assessment process.
• Document verification: Create a signed writing record that others can check.
• Batch analysis: Process multiple documents through repeated checks.
• API workflows: Integrate writing analysis into an organization's existing systems.

Why It Matters

AI writing detection can be difficult to interpret when a score is presented without information about how reliable that score is. A binary label can encourage people to treat an uncertain signal as definitive evidence, particularly in settings such as education, hiring, and internal review.

Know Nota takes a more cautious approach by displaying the false-positive rate alongside each result and abstaining entirely from scoring texts shorter than 100 words. It also does not claim that a text was definitively written by AI or attempt to identify a specific generator.

The platform adds another layer through source and writing records. By separating what was inferred, measured, matched, or observed, and by allowing signed records of how a document was written, it gives reviewers more evidence to consider alongside the analysis.

For organizations that need repeated checks, batch processing and API access extend the same evidence-focused workflow beyond individual documents.

Review Writing With More Context

Analyze documents for machine-like writing signals, examine source evidence, and create verifiable writing records without reducing the result to a simple yes-or-no label. With short-text abstention, false-positive reporting, document-level analysis, privacy controls, file review, and batch or API workflows, Know Nota provides a more cautious approach to writing assessment.