One of the nation’s most prestigious universities plans to reassess the very nature of higher education in the face of AI, a pervasive force that, if left unchecked, will undermine student development and the value of education itself.
AI is “upending foundational elements” of the Massachusetts Institute of Technology’s educational experience, a faculty-led committee found, warning that AI can credibly answer most written assessments across the undergraduate curriculum.
“Most classes taught at MIT will require careful review and, in many cases, substantial adaptation to survive an encounter with AI without serious disruption,” read the faculty report, which was composed after five months of meetings, research, and campus community outreach.
Increased AI use over the last three years has left students interacting less with peers and instructors, hampering social and cognitive development. Faculty uncertainty about AI has created confusion around classroom use, fueling temptation for students to rely on chatbots and complicating efforts to accurately assess competency.
MIT’s multi-faceted overhaul will focus on empowering humans to harness AI while mitigating its “disturbing risks,” President Sally Kornbluth said in a statement describing this “watershed moment for MIT.”
“It’s rare that a successful institution has to take a fresh look at many aspects of its mission,” she said. “Far and away the most important work will be cultural.”
The faculty committee proposed several recommendations for moving forward.
Rapidly retooling curriculum
MIT must streamline its curriculum approval processes to give departments greater authority to pilot AI-related changes quickly to keep pace with technology.
Departments should develop “AI light” and “AI heavy” courses and curricular pathways that prescribe the amount of AI use in the classroom.
More on AI readiness: Beyond ChatGPT: How to create successful students
Meaningful adaptation will require more than minor course revisions, the report contends. “Every subject taught at MIT will likely need to be reexamined and potentially revamped to make sure that how students are being taught, what they’re learning, and how they’re assessed are ‘AI-aware.'”
An “AI-aware” subject embraces the technology’s ubiquity in society and is led by instructors who provide clear expectations for its use, while recognizing that some students may still use AI inappropriately.
Rethinking grades and assessments
The report argues for increased in-person learning to strengthen student collaboration. Every course, regardless of size, should include structured activities like group projects, guided problem-solving sessions, peer feedback exercises and facilitated class discussions.
Experiential and project-based learning, oral exams, semester portfolios and out-of-class assignments must become central classroom components to deter students from relying on AI responses.
How students are evaluated on an assignment requires review as well. For example, the committee urged instructors to create multiple project deadlines for a single assignment, forcing faculty to shift their focus to students’ learning process rather than the final product.
The committee even suggested moving away from GPAs and grades so as to deter students from focusing on outcomes. “Rationing top grades would intensify the temptation to cut corners on actual learning by increasing reliance on AI,” the report read.
The committee also urges faculty to embed AI literacy and ethics across disciplines to educate learners on the technology’s societal and environmental impacts.
Establishing clear AI policies
The university will establish an institution-wide AI framework that enables departments to develop discipline-specific guardrails aligned with their academic objectives and instructional needs.
Guided by its department, each course will offer an AI policy. Instructors must explain why AI may, or may not, be used based on the course’s learning goals.
Ongoing training will help instructors better formulate their use and adoption of AI.
