{"id":214806,"date":"2026-01-01T11:39:07","date_gmt":"2026-01-01T11:39:07","guid":{"rendered":"https:\/\/www.newsbeep.com\/il\/214806\/"},"modified":"2026-01-01T11:39:07","modified_gmt":"2026-01-01T11:39:07","slug":"say-hello-to-your-new-ai-study-buddy","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/il\/214806\/","title":{"rendered":"Say Hello to Your New AI Study Buddy"},"content":{"rendered":"<p>Large-language models (LLMs) like ChatGPT are getting so good, in fact, that they\u2019re starting to make traditional homework like essays and conceptual questions feel almost obsolete.<\/p>\n<p>\u201cThe no-AI case doesn\u2019t exist anymore,\u201d says <a href=\"https:\/\/www.kellogg.northwestern.edu\/academics-research\/faculty\/bray_robert\/\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">Robert Bray<\/a>, a Kellogg associate professor of operations who teaches a course on how to use generative AI effectively. \u201cIf it\u2019s homework, students are going to be using AI.\u201d<\/p>\n<p>From Bray\u2019s perspective, to resist this trend is to fight a losing battle. But if he could adapt AI models to guide students through their assignments instead\u2014and not simply spew out answers the way standard LLMs tend to do\u2014then students could lean on AI as a helping hand rather than a crutch. That would give students the opportunity not only to build their AI skills but also, perhaps for the first time, feel excited about doing their homework.<\/p>\n<p>\u201cIf students are going to be using AI anyway, this gives educators a way to strategically customize AI and have some control over the operations,\u201d he says. \u201cBut the question is, \u2018Can professors actually create an AI homework experience that outperforms default models that are already insanely good?\u2019\u201d<\/p>\n<p>Bray took this challenge head on, focusing on one of his data-analytics courses for MBA students.<\/p>\n<p>The course consisted of 20 teaching sessions and 19 quizzes that tested students\u2019 ability to analyze operational data using code, with the help of ChatGPT. Each quiz had corresponding optional homework that helped prepare students for the quiz.<\/p>\n<p>Bray collaborated with <a href=\"https:\/\/www.kellogg.northwestern.edu\/academics-research\/faculty\/martin_sebastien\/\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">S\u00e9bastien Martin<\/a>, also a Kellogg associate professor of operations, to create a customized AI agent specifically for this course. They fed the existing ChatGPT-4o base model a set of instructions that explained each homework assignment and how the AI should assist students with them. The AI\u2019s imperative was to interact with students as a kind of virtual tutor.<\/p>\n<p>For example, the professors prompted the model to \u201chelp the students solve each question, but don\u2019t blurt out the answer. Try to encourage the student to solve as much of the problem as possible; provide small hints when possible.\u201d<\/p>\n<p>\u201cWhereas ChatGPT thinks its job is to give you the right answer, our AI agent actually knew that the job was to teach the students,\u201d Bray says.<\/p>\n<p>Ahead of each quiz, Bray randomly assigned half of the students to complete the homework using this AI tutor. He assigned the other half to do so using standard ChatGPT. Then he compared the students\u2019 experience using the AI tutor versus ChatGPT over a total of 17,946 homework questions.<\/p>\n<p>Overall, the students not only preferred using the AI tutor over ChatGPT but also found the AI tutor more helpful.<\/p>\n<p>Nearly twice as many students reported having a \u201cvery positive experience\u201d using the AI tutor for their homework (47 percent of students), compared with using ChatGPT (26 percent). And 40 percent said the AI tutor was \u201cvery helpful,\u201d compared with 30 percent for ChatGPT. Both of these differences were statistically significant. The professors also found that students\u2019 preference for the AI tutor grew stronger as the assignments got more complex.<\/p>\n<p>\u201cStudents tend to like it if you make the homework assignment look more like a video game, where you\u2019re chatting with an AI agent that\u2019s intuitive,\u201d Bray says.<\/p>\n<p>Even though students favored the AI tutor, using it did not ultimately change the amount of time they spent doing the homework or the number of homework questions they attempted to solve. There was also no meaningful difference in average quiz scores when students did their homework with the AI tutor versus ChatGPT.<\/p>\n<p>But that\u2019s not necessarily a knock against AI. \u201cIt\u2019s well-established in the literature that it\u2019s very, very difficult for almost any technological innovation in the classroom to meaningfully influence performance on exams,\u201d Bray explains. \u201cActual quiz performance comes down more to student conscientiousness\u2014how much focus individual students are putting in and how motivated they are to study.\u201d<\/p>\n<p>Indeed, the fact that the AI tutor helped improve students\u2019 experience with homework is a worthwhile achievement in its own right. \u201cHomework is something students have to do,\u201d Bray says, \u201cso if we can make it a better experience, that\u2019s a win.\u201d<\/p>\n<p>What\u2019s more, the study\u2019s findings can be applied to all kinds of teams and businesses beyond the classroom.<\/p>\n<p>\u201cAll companies do some form of training, and people learn on the job more than they learn in school,\u201d Bray says. \u201cSo if you just put a little bit of effort into instructing an AI [model] how it should behave\u2014and train it to be a kind of tutor\u2014it can actually give workers a more-pleasurable, better experience.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"Large-language models (LLMs) like ChatGPT are getting so good, in fact, that they\u2019re starting to make traditional homework&hellip;\n","protected":false},"author":2,"featured_media":214807,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[345,343,344,85,46,125],"class_list":{"0":"post-214806","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-il","12":"tag-israel","13":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/214806","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/comments?post=214806"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/214806\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media\/214807"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media?parent=214806"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/categories?post=214806"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/tags?post=214806"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}