Science & Tech · AI · 14 hrs ago
UC Berkeley professor shifts homework grading as students use AI
At UC Berkeley, mathematician Richard Borcherds has stopped grading homework for correctness after seeing students use AI tools to earn high marks.
He now gives credit mainly for completing assignments, while acknowledging that this is not a good solution.
Borcherds says AI makes it difficult to tell students’ work from machine-generated answers, and paying graders to mark those answers would waste time and money.
He says in-person exams without electronic devices now carry more weight in his grading, while online exams are less useful because students can use AI.
The change reflects a wider problem for universities trying to assess learning as AI tools become more capable.
Borcherds also worries that AI could outperform human mathematicians in the coming years, though he says he does not know whether students should still pursue mathematics.
UC Berkeley leaves AI rules to instructors, asks them to state course expectations in syllabi, and provides sample policies; the campus has not adopted AI detection tools as a standard.
UC Berkeley mathematics professor Richard Borcherds has stopped grading homework mainly for correctness because students can use AI to earn high scores.
He now gives credit primarily for completing assignments, while acknowledging that he has no better system.
Borcherds says AI-generated work has become difficult to distinguish from students’ own work.
He bases grades mainly on in-person exams, where students cannot use electronic devices.
UC Berkeley requires faculty to address AI use in course expectations, but policies vary by course.
- Who
- Richard Borcherds, a UC Berkeley mathematics professor and Fields Medalist.
- What
- He has shifted homework grading from correctness to completion because of widespread student use of AI.
- When
- Not stated. The Daily Cal reported on the change on October 23, 2026.
- Where
- UC Berkeley.
- Why
- Borcherds says grading for correctness can encourage students to use AI and penalize those who work honestly.
This story does not have two clearly opposing sides.
Borcherds discussed AI and homework grading in an interview with the Daily Cal.
The Daily Cal reported that Borcherds grades homework based on completion rather than correctness.
Yahoo published an account of Borcherds’ grading approach and his concerns about AI in mathematics.
- Professor
- Richard Borcherds
- University
- UC Berkeley
- Homework grading
- Credit based primarily on completion
- In-person exams
- Students are not allowed to use electronic devices
- AI policy
- Faculty are asked to state course expectations about AI use; policies vary by course
- AI detection tools
- Campus tried Turnitin and Rumi but found them inadequate for campus-wide standardization
3 sources
What people are saying
Grading for correctness just adds a huge incentive for them to use AI for homework instead of doing it themselves, and penalizes students for honesty.
If AI continues to improve at present rate, then, in a few years, there’ll be nothing human mathematicians can do that AI can’t do better.
I have no idea what to advise. I even asked ChatGPT this exact question, and even ChatGPT said it didn’t have any idea what was going to happen.
There is no easy way to tell the difference between human and AI generated work.