AI-1187 - USA (South Carolina) - Virtual Testing AI Monitoring and Proctoring Services - Deadline August 24,2026

(1) Vendor needs to provide virtual testing AI monitoring and proctoring services.– Provide a secure, scalable, and zero–install virtual testing AI monitoring and proctoring solution. – The ideal solution will provide AI–powered automated behavior monitoring, real–time integrity tracking, and optional human/hybrid review capabilities, fully integrated within our existing educational technology ecosystem.A. Core Platform Integration & Environment- LMS Integration: Native integration with Canvas LMS via LTI standards.- Single Sign–On (SSO): Full compatibility with ClassLink for automated rostering and single sign–on access.- Hardware & OS Compatibility: Primary optimization for Windows, as well as, macOS and Chromebooks (ChromeOS) devices.B. 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).C. 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.(2) All the questions must be submitted no later than August 21, 2026.