Editor AI Guard

Here's the elevator pitch What's your idea? This will be a short tagline for the project

Comment

Inspiration

With universities rapidly deploying blunt AI detectors, countless honest students and international researchers—particularly ESL (English as a Second Language) writers—are facing false-positive accusations simply because formal academic phrasing statistically resembles AI patterns. Meanwhile, institutional checkers like Turnitin remain locked behind expensive enterprise walls. We built Editor AI Guard to restore fairness and provide transparent, auditable academic defense.

What it does

• Multi-Engine AI Detection: Transparent Perplexity, Burstiness variance, and Entropy calculations with sentence-by-sentence highlights. • Academic Sentence Humanizer: Ethically restructures mechanical phrasing while rigorously preserving scientific citations and research tone. • Live Scholarly Plagiarism Check: Real-time scanning against 150M+ peer-reviewed DOIs via Crossref. • Cryptographic QR Verification: Issues verifiable, tamper-evident authorship certificates to dispute false-positive AI flags. • Zero Data Retention: Privacy-first architecture that never stores or trains on submitted research. • Native localization across 13 international languages.

How we built it

Built on Cloudflare Workers for sub-50ms global latency and zero data leakage. The frontend utilizes lightweight vanilla JavaScript with modern CSS glassmorphism, paired with custom client-side statistical linguistic algorithms and Crossref REST API indexing.

Built With

Share this project:

Updates