artefakt

Artefakt – process-aware academic integrity for Microsoft Teams for Education

A similarity score isn't a verdict, and a snapshot of a finished submission isn't a story. To support a student who is genuinely struggling, and to recognise when something else is happening, you need context, not an isolated number.

Artefakt analyses every revision of a student's work as it is written, so you see how each piece of work came together: matching text, AI-generated patterns, and significant paste events tracked across the life of the document. You attach artefakt to an assignment in one click and see a class-wide view of the writing process for every student. Outliers are surfaced. Every finding is traceable back to its source. The decision always stays with you.

The same view that helps you confirm work is genuinely a student's also helps you spot a student who is struggling. Artefakt informs your feedback in formative tasks and gives you confidence in summative ones.

Built for teachers, instructional designers, and academic integrity officers at institutions running Microsoft Teams for Education.

Key benefits

• Assessment grounded in evidence, not a single score
• Earlier visibility into students who need support
• Defensible decisions on academic integrity, traceable to source

Capabilities

• One-click attachment to any Microsoft Teams assignment
• A revision-level view of how each piece of work was written
• Text similarity matching with privacy-compliant source attribution
• AI-generated content pattern detection
• Significant paste event detection
• A class-wide overview that surfaces what's worth a closer look

Requirements

• Microsoft 365 Education tenant
• Installation by a Microsoft 365 Global Administrator (one-time tenant consent for Microsoft Graph Education permissions)
• Teacher (faculty) licence required to view class results
• Active artefakt subscription

AI use and reporting

Artefakt uses AI to detect patterns associated with AI-generated content and flag them for your review. AI surfaces findings; you always make the decision. AI-generated content detection ships with Swedish-language support enabled. Coverage for additional languages is available on request. Text similarity, paste-event tracking, and revision analysis are language-agnostic.

To report objectionable AI-generated content, email support@etos.org.

Sign up and support

Sign Up: contact us at https://www.etos.org/contact or contact@etos.org

Get Started: https://www.etos.org/guides

Help / Support: support@etos.org