Virtual Testing AI Monitoring and Proctoring Services

The Vendor is required to provide AI-powered automated behavior monitoring, real-time integrity tracking, and optional human/hybrid review capabilities, fully integrated within our existing educational technology ecosystem.
- Core platform integration & environment
• LMS integration: native integration with canvas LMS via LTI standards.
• Single sign-on (SSO): full compatibility with class link for automated rostering and single sign-on access.
• Hardware & OS compatibility: primary optimization for windows, as well as, macos and chrome books (chromeos) devices.
• Zero-install architecture: the proctoring system must run 100% inside the browser without requiring students to download executable software, local apps, or third-party browser extensions.
- AI behavior monitoring & security capabilities
• Identity verification: automated pre-exam photo id verification and continuous facial recognition throughout the assessment session.
• Real-time AI behavior detection: automated detection of multi-face presence, gaze/eye-tracking deviation, secondary device usage, ambient speech/audio detection, and off-screen focus shifts.
• AI & LLM detection: automated mechanisms to detect and prevent unauthorized usage of external AI tools, large language models (LLMS), or web searches during testing.
• Exam environment control: built-in browser lockdown functionality (restricting tabs, copy-paste, print screen, and virtual machines).
• Secondary camera support: capability to leverage secondary mobile device cameras for 360-degree room scans or secondary monitoring angles when required.
- Admin dashboard & hybrid review
• Session length flexibility: support for varying exam windows ranging from 30 minutes to 3 hours.
• Flagging & review dashboard: time-stamped event logging with video/audio clips, flag severity ratings, and an objective integrity score for fast administrative review.
• Hybrid / human review option: capability for flagged sessions to be escalated to trained human reviewers or live proctors to minimize false positives and support academic decisions.
• Ethical AI & transparency: transparent flagging algorithms that serve as evidence for educator review rather than automated auto-failing of students.