New: exact-layout reports — annotations directly on the original document  See what's new
Similarity · AI-writing analysis · Verifiable certificates

Academic integrity your institution can defend ✓ verifiable

iOriginally brings similarity detection, transparent AI-writing analysis, and independently verifiable certificates into one platform — trusted by examiners, faculty, and administrators to produce answers that stand up to scrutiny.

14-day institutional evaluation Cloud or self-hosted GDPR-ready
Built for high-stakes academic decisions
Examination offices Graduate schools Research institutes Accreditation teams
Experienced educator with an open book, glasses in hand

Works with the systems your institution already runs

<2s
Exclusion filters recompute the score live — no re-analysis queue
800
Pages per document, with smooth virtualized viewing
9+
File formats accepted, plus ZIP batches for thesis season
Zero
Fabricated results — every number is computed from evidence
How it works

From submission to defensible answer

1

Submit, any way you work

Students submit through your LMS, offices bulk-upload ZIPs of theses, systems integrate over the API — nine file formats, virus-scanned on arrival.

LMS launchBulk ZIPREST API
2

Analyse in three layers

Similarity against internet, publications, and your repository — plus AI-writing signals and byte-level integrity flags, each reported separately.

SimilarityAI signalsIntegrity flags
3

Decide with evidence

An interactive report with live filters, an annotated PDF on the original layout, and a certificate anyone can verify by QR — audit-logged end to end.

Live filtersCertificatePublic verification
The similarity report

One report. Every answer your committee needs.

Highlighted matches with named sources, side-by-side comparison, and live exclusion filters — designed for the moment a result is challenged.

app.ioriginally.com/report/9f2c-e871

Side-by-side evidence

Click any highlight to see submission and source aligned — exact matches and paraphrases as distinct layers.

Live exclusion filters

Quotes, bibliography, small matches, and assignment templates — the score updates in under two seconds.

Annotated on the original

Downloadable reports highlight matches directly on the uploaded document — layout, pagination, and fonts preserved.

Detection, in depth

Three layers of scrutiny, reported separately

Similarity, AI-writing signals, and document-integrity flags are analysed and presented as distinct layers — so reviewers always know exactly what kind of evidence they are looking at.

Similarity detection

Matching against internet sources, scholarly publications, and your institution's own repository — with paraphrase detection and cohort-level collusion analysis built in.

Internet & publicationsParaphrase layerCollusion

AI-writing analysis

Segment-level probabilities, highlighted in place and kept strictly separate from the similarity score. Every AI screen carries your institution's policy note — a signal, never a verdict.

Per-segment evidencePolicy noteSeparate layer

Integrity flags

Hidden white text, look-alike Unicode characters, zero-width spacing, and image-only pages — detected at the byte level during ingestion and surfaced with location evidence.

Hidden textHomoglyphsOCR fallback
Verifiable certificates

Certificates anyone can verify in seconds

Every completed analysis can issue a formal, institution-branded originality certificate with a QR code that resolves to a public verification page. Employers, examiners, and partner universities can confirm authenticity instantly — no login, no document exposure.

  • Forgery ends at the scan. A tampered PDF fails the public check the moment anyone verifies it.
  • Mapped to your policy framework. Configure similarity bands to your national standard or your own thresholds.
  • A receipt for every submission. Verification code, SHA-256 fingerprint, timestamp — recorded and emailed automatically.
See it live in your demo
Graduates in gowns celebrating in front of a university building
Senior professor with a laptop in a university lecture theatre
Bulk thesis checking

Built for thesis season

Research offices and examination teams check submissions in bulk, not one at a time. Quick Submit takes a ZIP of two hundred theses — no course setup required — and returns a working table of scores, policy levels, flags, and one-click certificates.

  • Per-file metadata — candidate, supervisor, department, and degree flow straight onto each certificate.
  • CSV export for the record — results land in the format your office already files.
  • Versioned resubmissions — a revised thesis produces a new report you can diff against the last.
Walk through Quick Submit
Solutions by role

One platform for every desk

Integrity officers, instructors, students, and administrators each get a purpose-built workflow — not a shared screen with different permissions.

Bulk checking without course setup

Quick Submit handles individual theses or ZIP batches of hundreds, with metadata that flows to certificates automatically.

Results as a working table

Scores, policy levels, flags, and certificate actions in one sortable view — exportable to CSV.

Defensible months later

Every download, decision, and certificate is audit-logged, so challenged cases are settled with records, not recollections.

Quick Submit · thesis batch · June
DocumentSimilarityAIStatus
M. Chen — Carbon pricing…24%8%Certified
A. Okafor — Grid resilience…11%4%Certified
L. Nováková — Protein folding…38%12%Review
S. Haddad — Urban hydrology…52%6%Review
J. Park — Semantic parsing…78%61%Escalated
198 of 214 complete · export CSV

An inbox that reads at a glance

Similarity and AI chips, integrity flags, lateness, and grades in sortable columns, with bulk actions for downloads and resubmission requests.

Feedback where the text is

Inline comments, reusable QuickMarks, weighted rubrics, and voice notes — grading lives next to the evidence and releases on your schedule.

Template text excluded automatically

Upload the assignment brief once — boilerplate is fingerprinted and excluded from every score in the class.

ECON 402 · Essay 2 · due 18 June
StudentSimilarityAIGrade
E. Johansson9%2%A−
D. Mensah17%5%B+
R. Nakamura31%9%
T. Álvarez 2 flags44%28%
18 submitted · 3 late · 1 flagged

Watch the check happen

Drag, drop, and follow a live pipeline — scan, extract, match, report — instead of staring at a spinner.

Proof of submission, always

A digital receipt with verification code and file fingerprint, emailed instantly — disputes about lost uploads end here.

Learn from the report

Where the instructor allows it, students see their own report — citation gaps become a lesson before they become a case.

Submission pipeline · live
Virus scan & file validation
Text extraction · 214 pages
Similarity & AI analysis~40s
Report ready
Receipt RCP-2026-8841 emailed · SHA-256 recorded

The whole institution, one dashboard

Usage, similarity trends, policy-level distributions, and most-active departments — exportable for accreditation and annual review.

Your policy, brand, and data

Institution-wide defaults, your logo on every report and certificate, retention windows you control, and bulk user import with dry-run validation.

Procurement-friendly billing

Annual licenses or prepaid credit packs, quoted per institution — purchase orders and bank transfer are first-class payment paths worldwide.

Institution analytics · spring term
4,812
submissions checked
18.3%
median similarity
Similarity distribution
0–1010–2020–3535–5050–7575+
audit log · 30,412 events · export XLSX
Security & data stewardship

Student work is unpublished IP. We treat it that way.

Custody, isolation, and deletion are engineered to be provable — not promised. Run iOriginally in the cloud, or entirely on infrastructure your institution controls.

Database-enforced isolation

Row-level security on every tenant table — institution boundaries are enforced in the database itself and covered by automated cross-tenant tests.

Never used for training

Student documents are never used to train models and never shared across institutions unless you explicitly opt into the shared repository.

Retention you control

Per-institution retention windows with automated purge, legal-hold flags, and full export or deletion on request — GDPR- and DPDP-ready.

Hardened by default

Antivirus scanning on every upload, signed expiring file URLs, SHA-256 fingerprints, modern password hashing, and 2FA for administrative roles.

Everything is audited

Logins, admin actions, report downloads, certificate issuance — a filterable audit trail your compliance office can query years later.

Self-hosted option

The full platform — detection engines included — deploys on servers you control, making data residency a configuration choice, not a negotiation.

GDPR-ready DPDP-ready WCAG 2.1 AA SSO / OIDC & 2FA Encrypted in transit & at rest
How we compare

An honest comparison — including where we don't lead

We sell with the truth. Row by row against the incumbents, including the one place the market leader is still ahead.

CapabilityiOriginallyTurnitinDrillBit
Live filter recompute — no re-analysis~
Anti-masking flags — hidden text, homoglyphs, image-only pages~
Public certificate verification — QR, no login~
Report annotated on the original document layout~~
Fully self-hosted deployment option
Licensed closed-publisher full-text corpus~
The honest bottom line: Turnitin still leads on its licensed publisher corpus — and we say so on our own website. iOriginally leads where the daily work happens: verifiable certificates, anti-masking defense, live filters, self-hosting, and flat institutional pricing that doesn't scale per seat.
From integrity teams

Trusted where the decisions are hard

An external examiner questioned a certificate. We scanned the QR code, the verification page confirmed it in seconds, and the matter was closed before the meeting began.
Portrait of Dr. Elena Marsh Dr. Elena Marsh
Director of Research Integrity · public research university
We moved two hundred theses through Quick Submit in an afternoon. Filters recompute instantly, so review meetings stopped being arguments about bibliographies.
Portrait of Prof. Marcus Ellery Prof. Marcus Ellery
Dean of Graduate Studies · comprehensive university
The AI report shows its segments and says "signal, not proof" right on the page. That one sentence keeps our academic panels calm and our decisions defensible.
Portrait of Amara Osei Amara Osei
Head of Academic Standards · international university
FAQ

Frequently asked questions

Can we run iOriginally entirely on our own infrastructure?
Yes. The full platform — including the detection engines — deploys on servers your institution controls, and a mid-size university runs comfortably on a single well-specified machine. Managed cloud hosting is available if you'd rather not operate it yourself.
Are student documents used to train models or shared with other institutions?
Never for training, and never shared across institutions unless you explicitly opt into the shared repository. Retention windows, export, and deletion remain under your control, with GDPR- and DPDP-ready processes.
How reliable is the AI-writing detection?
We treat it as what it is: a probabilistic signal. iOriginally shows per-segment probabilities, keeps the AI number strictly separate from similarity, and prints your institution's policy note on every AI screen and PDF. We will never present it as proof — and we put that in writing.
Does it work with our LMS and student information system?
iOriginally is LTI 1.3-native, so it launches from Moodle, Canvas, Blackboard, Brightspace, and any other compliant LMS with single sign-on and automatic role mapping. The REST API and signed webhooks cover SIS and ERP integration.
Can similarity thresholds match our national regulations?
Yes. Policy bands are configurable per institution — map them to your national framework or define your own thresholds, and they appear consistently on reports, certificates, and analytics.
How long does implementation take?
The LMS connection is typically configured in an afternoon — LTI 1.3 registration plus single sign-on. Most institutions run their first real thesis batch during the 14-day evaluation, before any contract is signed.
Can we switch from Turnitin or DrillBit mid-contract?
Yes — most institutions run iOriginally in parallel during the evaluation period and cut over at term boundaries. Your institutional repository builds from your own submissions from day one, and our team assists with migration planning.
What support and training is included?
Every license includes onboarding for administrators and the integrity office, role-based training materials for faculty and students, and email support. Annual licenses add priority support with guaranteed response times.
How does purchasing work for a public institution?
Quote-led and procurement-friendly: we quote to your institution's size in your currency, accept purchase orders and bank transfer as first-class payment paths, and issue tax-compliant invoices for your jurisdiction. Card payment is available but never required.

See iOriginally on your own documents

A 25-minute walkthrough with your integrity team — real submissions, real reports, real certificates. No slide decks, no commitment.

Bring your own theses — we'll run them live during the call See the exact workflow for your LMS and policy framework Leave with a 14-day evaluation tenant, no credit card

Book your demo

Our team replies within one business day.

By submitting, you agree to our privacy policy. Your documents are never used for training.

Request received

Thank you — our team will reach out within one business day to schedule your walkthrough.