In academia, using AI for grading and providing a feedback seems like a reasonable next step. I suspect that AI will manage to grade and provide valid feedback for easy to medium level of difficulty calculation problems. If it can solve it, it can grade it, right? What if the problem is not a calculation assignment? We need to test it! Note, the privacy of my students is a major concern for me (even with the turn off button that instructs LLM not to save the information; a
