AI Supported Grading integrates the Inspera Graide (IG) engine directly into the Marking 2.0 workspace to provide scalable, consistent, and faster evaluation of open-response questions.

Key Principles

• Human-in-the-Loop: The AI acts as an assistant. Graders retain full authority and must review, confirm, or override all AI-generated marks before they are finalized. There is no fully autonomous grading.
• Supported Question Types: Text Area (with single interaction blocks) and Essay Type questions.
• Model Training: The AI model requires the grader to evaluate the initial submissions manually. Once enough submissions are graded to build model confidence, the AI will begin generating automated suggestions for subsequent responses.
• Marking 2.0 Only: AI assistance is exclusively available within the Marking 2.0 interface. It is not compatible with Marking 1.0.