MIT Issues Urgent Call to Restructure Higher Education as Generative AI Completes Most Undergraduate Work

AI can now credibly complete most undergraduate assignments, MIT warns

The MIT campus in Cambridge, Massachusetts. (Patrick Gillooly/MIT) Generative artificial intelligence is advancing so rapidly and producing such massive, long-term disruptions to education that a committee at MIT proposed profound changes Tuesday to counter the technology's risks. In a message to campus sharing the committee's report, MIT President Sally Kornbluth called the opportunities and risks of AI a "watershed for MIT -- and for all of higher education." The report found that even as AI accelerates and expands discoveries, it is also driving dramatic changes in campus culture, upending foundational experiences such as study groups, office hours and undergraduate research. The technologies are now able to credibly complete most undergraduate assignments, wrote the committee, co-chaired by professors Eric Klopfer and Samuel Madden. Most classes at MIT should be reviewed, and many will need substantial changes, they wrote, urging instructors to consider other means of assessment -- including oral exams, portfolios and in-person conversations tied to work done outside class. The report called on university officials to make it easier and quicker to modify courses, to provide more physical spaces for students and faculty to meet in person, and to build community. "For the sake of our students, for the future of MIT and for the shape of our sector and our society," Kornbluth wrote, "it's imperative that we get this right." Catch up quickly every weekday morning with a rundown of the 7 most important and interesting stories. Sign up for The 7 newsletter.

MIT says AI requires a rethink of how students learn

MIT said generative AI requires a broader overhaul of teaching, learning and research training, moving beyond classroom rules to changes in assessment, course design and student skill development. The institute's Ad Hoc Committee on AI Use in Teaching, Learning, and Research Training said in a report released on Thursday that MIT should reconsider what students learn, how faculty assess learning, where AI belongs in that process and where it should be excluded. President Sally Kornbluth called the moment a "watershed" for MIT and higher education, and wrote: "This is not an optional exercise." The committee was formed in January 2026 with three assignments: assess AI use by faculty and students, identify new approaches to teaching and assessment, and propose an AI use policy. Its final report expanded that mandate to ask what an MIT education should mean when AI can write, code, summarize, analyze and simulate at rapidly improving levels. MIT proposed three broad responses. Courses and programs should become AI-aware, the residential experience and human community should receive greater attention, and the institute should create permanent mechanisms for experimentation and policy revision. One recommendation calls for every class to state clearly whether students may use AI, must use AI for specific work or must avoid it entirely. The report gave examples including writing seminars that allow AI for critique but not initial drafts, computer science courses that require students to build basic algorithms unaided before using coding agents, and laboratory courses that permit AI analysis but still require physical experiments and in-person defense of results. MIT's Teaching and Learning Lab has already published course examples using that approach. In one language course, students produce their own translation, compare it with an AI-generated version and then analyze the machine's choices, while a data visualization assignment asks students to compare an AI-assisted attempt with one developed through instructor guidance. The report said assessment may need substantial rebuilding because finished essays, answers and projects may no longer show what students know or can do independently. Stanford's Accelerator for Learning and ETS said in July, after a convening of more than 100 education, research and policy leaders, that schools should use portfolios, conversations, performance tasks, formative feedback and demonstrations of competence to provide richer evidence of learning. The University of Sydney has adopted what it calls a two-lane assessment model. One lane uses secure, often in-person work to verify what students can do independently, while the second allows relevant tools, including AI, so students learn how to use them in realistic settings. Research cited in the article points to mixed outcomes depending on how AI is used. A March 2026 meta-analysis of 35 experimental studies covering 4,193 participants found a moderately positive overall effect from ChatGPT use on learning outcomes, while a separate 2026 review of 67 studies found AI could support critical and creative thinking when instructors built it into structured inquiry, reflection and evaluation, but found signs of cognitive offloading in loosely structured settings. MIT also said it wants to protect the value of residential education and human relationships as information, tutoring, coding assistance and drafting become cheaper and more abundant through AI tools. The report said laboratory sessions, oral defenses, team projects and in-person problem solving provide evidence of how a person thinks that AI cannot easily certify. The committee said AI tools are already helping researchers generate hypotheses and test possible solutions faster, and MIT plans discipline-based communities to help faculty use such systems responsibly in research.