Privacy Policy
Effective Date: October 18, 2026 · Version 2.0
1. The Zero-Vault Guarantee
Unlike conventional plagiarism detectors that index submitted manuscripts into a persistent global database (where subsequent drafts of your own paper may trigger false similarity flags), ScholarScreen operates on a strict Zero-Vault ephemeral memory model.
Your document text is held exclusively in volatile RAM during active pipeline execution. Once the forensic screening finishes and your verification report is produced, the memory buffers are wiped. No manuscript is ever saved, indexed, archived, or shared.
2. No AI Training on User Submissions
We firmly believe academic authors and students retain absolute ownership of their intellectual property. ScholarScreen never utilizes submitted text, drafts, or manuscripts to train, fine-tune, or calibrate any machine learning, generative, or neural models.
3. How Forensic Probing Works
To detect textual overlap and synonym drift, the engine extracts distinct n-gram shingles and queries public academic registries (Crossref, Semantic Scholar, OpenAlex, Europe PMC, and open preprint indices). Only concise search phrases are sent to these public discovery APIs—your complete manuscript is never broadcast.
4. Payments & Access Voucher Codes
Financial transactions for Student Plans (₹99), Author Packs (₹499), and Campus Pro (₹49,000) are processed through Razorpay over encrypted TLS channels. ScholarScreen never receives or stores your credit/debit card numbers, CVVs, or UPI credentials.
Vouchers and Campus keys are managed using cryptographic hashes strictly to verify remaining quota balances and enforce single-use or daily rate caps.
5. Tamper-Proof PDF Certificates
Every completed screening generates a verification certificate bearing a cryptographic SHA-256 seal. This certificate summarizes numeric scores, matched sources, and timestamps for your committee defense, while keeping the manuscript contents private to the author.
Contact Privacy Governance
For technical inquiries regarding our memory lifecycle, ephemeral execution, or data governance, reach out to our team at [email protected].