{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/d7085e9f5c6a43bab780dd757c672e33\" frameborder=\"0\" width=\"1664\" height=\"1248\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1248,"width":1664,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1248,"thumbnail_width":1664,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/d7085e9f5c6a43bab780dd757c672e33-9c0151f17258ebd3.gif","duration":95.1186,"title":"Calibration","description":"In this video, I discuss the process of reviewing AI grades and the potential misalignment between AI grading and our established rubric standards. I highlight that if you notice repeated discrepancies after grading 10 to 20 samples, you can click on Edit Rubric and then Generate Colors. This action will provide a detailed and granular description of each rubric item, clarifying how the AI should approach grading. It's essential to ensure that our grading criteria are consistently applied. Please take this step if you encounter any issues with AI grading."}