跟进：加州大学伯克利分校菲尔兹奖得主不再按正确率批改作业：称人工智能使传统作业评分失去意义，改以完成情况计分
伯克利消息：据《纽约邮报》报道，加州大学伯克利分校数学教授、1998 年菲尔兹奖得主理查德·博彻兹（Richard Borcherds）已停止按正确与否批改学生作业，原因是学生广泛使用人工智能工具已使传统作业评分失去意义。
博彻兹告诉《加州邮报》：“学生用人工智能做作业现在已经很普遍。按正确率评分只会大大刺激他们用人工智能做作业而不是自己做，还会惩罚诚实的学生。”他还说，付钱请批改员评估人工智能生成的答案也是浪费时间与金钱。博彻兹现在主要根据学生是否完成作业给予学分，而不是答案是否正确，并承认自己不知道更好的办法。
他表示，如今已很难区分人工智能产出的作业与学生自己写的作业：“没有简单办法区分人类与人工智能生成的作业。偶尔能看出是人工智能，因为人工智能犯了几个特征性错误，但人工智能现在进步如此之快，很少再犯明显错误。”博彻兹还表示，他现在主要依据不允许使用电子设备的现场考试评分；线上考试则成了“浪费时间”，因为一些学生用人工智能答题，而且他注意到线上考试的平均分大幅上升。
加州大学伯克利分校已制定指导方针，要求教师在课程中说明人工智能使用问题；伯克利学术评议会分部要求教师在教学大纲中加入相关说明，包括管理人工智能工具的课程规则。学校曾探索 Turnitin 与 Rumi 等人工智能检测软件，但认定这些工具不够可靠，不足以作为全校统一标准采用；各教师仍自行掌握课堂人工智能政策。
UC Berkeley Professor and Fields Medalist Stops Grading Homework for Correctness: Richard Borcherds Says AI Has Undermined Traditional Homework Grading and Now Awards Credit for Completion
BERKELEY — A UC Berkeley professor has stopped grading students' homework for correctness after noticing artificial intelligence tools were helping them rack up top marks, the New York Post reported. Richard Borcherds, a mathematician who received the prestigious Fields Medal in 1998, said widespread student use of AI has undermined the value of traditional homework grading.
"Student use of AI for homework is now widespread," Borcherds told The California Post. "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." Paying graders to evaluate AI-generated answers is also a waste of time and money, he said. Borcherds now awards credit based primarily on whether students complete their assignments rather than whether their answers are correct, acknowledging that he does not know of a better solution.
"There is no easy way to tell the difference between human and AI generated work," he told The California Post. "Occasionally one could tell it is AI because AI made a few characteristic errors, but AI has now improved so much that it rarely makes obvious mistakes." Borcherds told The California Post that he now bases grades mainly on in-person exams, where students are not allowed to use electronic devices. Online exams, meanwhile, have become "a waste of time," he said, because some students use AI to answer the questions. He cannot tell which students are using AI but has noticed that average scores rise substantially on online exams.
UC Berkeley has established guidance requiring faculty members to address AI use in their courses; the Berkeley Division of the Academic Senate asks instructors to include statements in their syllabi outlining course expectations, including rules governing AI tools. The university has also explored AI-detection software, including Turnitin and Rumi, but determined the tools were not reliable enough to adopt as a standardized campus-wide solution, the Daily Californian reported. Individual instructors therefore retain control over AI policies in their classrooms.
来源：GSM·美国（原文链接）
图片来源：图：加州大学伯克利分校校园，自萨瑟塔台阶向西望（资料图）。照片：Firstcultural，CC0，经维基共享资源。 / The UC Berkeley campus, looking west from the steps of Sather Tower (file photo). Photo: Firstcultural, CC0, via Wikimedia Commons.
