- Math homework, dominated by numbers, plays to AI strengths and is far harder to detect than AI-written prose
- Caltech has no unified AI policy, while some courses explicitly allow AI use
- Students argue that leaning on AI for homework backfires in closed-book in-person exams
AI CURATED HOT STORY · 11/17/2026
Caltech Professor Drops Grading for Math Homework as AI Takes Over
Caltech mathematics professor Yi Ni has announced that he will no longer grade homework based on correctness, but only check whether students completed it. The decision stems from the widespread use of generative AI, which excels at solving math problems, making it difficult for professors to distinguish AI-generated work from student work. Caltech has no unified AI policy, and some courses explicitly allow AI use. Students acknowledge that relying on AI for homework will backfire in closed-book exams, which remain the primary assessment. The move shifts evaluation toward in-class exams and has led some students to reconsider majoring in math. This case highlights universities' passive adaptation to generative AI, as traditional homework-based assessment loses effectiveness in math, prompting a reevaluation of evaluation systems and academic integrity in the AI era.
Hourly heat curve
23 fully observed hoursThe line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.
Coverage timeline
Total 2 reports · Latest first- IT之家 · 智能时代T2·58 pts
- Hacker News AIT2·58 pts
- Homework no longer reliably distinguishes student ability, pushing weight toward in-class exams
- Caltech has no unified AI teaching guidance, leaving policies to individual courses
- Rising AI capability is reshaping undergraduates' decisions to major in math