Harvard College Dean David J. Deming told incoming first-year students that higher education is entering a "third era" driven by artificial intelligence, arguing that AI will commodify expertise and force universities to become more communal. His remarks came during an abbreviated Convocation ceremony for the Class of 2030 held Tuesday afternoon in Sanders Theatre.
Deming framed the history of universities in two phases. In the 1600s, Harvard's value stemmed from its physical collection of books, a rare resource when information access was tightly restricted. As knowledge became widely available, the role shifted toward cultivating specialized expertise, a model Deming said often fractured common intellectual ground.
"Specialization has great benefits, but also great costs because it divides us into separate realities and inhibits the formation of common firms. I believe this second era is coming to a close," Deming said.
The commodification of expertise
Deming pointed to AI's ability to personalize learning in ways that recorded lectures or open internet content cannot. He distinguished between static expert material and interactive tools that adapt to individual questions and needs. AI tutors, he noted, have already demonstrated improved student learning outcomes.
"The internet has made expert content available to anyone in the world," Deming said. "That content doesn't replace your education here because it isn't personalized. Masterful lectures are part of your education, but not the only one. They don't answer your questions or meet your individualized learning needs." He added: "But artificial intelligence can be fully personalized."
The long-term implication, Deming argued, is that expertise itself becomes a commodity. This shift does not render universities obsolete, but it does change what they must offer. AI for Education is reshaping how institutions approach that challenge.
A push toward communal learning
Deming directly challenged the notion that technological access eliminates the need for college. He cited tech leaders who encourage students to skip higher education, calling that view a misunderstanding of a university's purpose.
"Tech leaders like Elon Musk encourage you to skip college, saying things like, 'You don't need to go to college to learn stuff; everything is available,' basically. But this misunderstands what a university is for," he said.
Instead, Deming said universities should lean into strengths that technology cannot replicate. He described a future where undergraduate education becomes "radically more communal," focusing on collaborative skills and forms of learning where technology offers little help. This aligns with his previously stated view that social skills - collaboration and effective communication - gain value as AI handles more technical tasks.
Resources like AI for Teachers reflect the growing need for educators to understand where AI fits and where human-led, communal learning remains essential.
A split ceremony and student response
Rain forced the event indoors to Sanders Theatre, whose limited capacity split the class into two identical, abbreviated ceremonies. All musical performances were cut to accommodate the schedule.
Some students welcomed Deming's direct engagement with AI. Kabir S. Mahajan '30 said the speech acknowledged "the elephant in the room" while articulating what Harvard offers beyond what technology can provide.
Deming has made AI a priority during his tenure as College dean. At an April open forum, he said he hoped to "push harder" on the technology, though he has resisted a blanket AI policy for courses, preferring that instructors set limits on a course-by-course basis.
Why this matters for educators
Deming's argument signals a practical shift for anyone working in education: if AI commodifies the delivery of expert knowledge, then the classroom's value moves toward what happens between people. Lecturing alone won't differentiate an institution. Designing learning experiences that require collaboration, debate, and real-time interpersonal problem-solving will. For teachers and administrators, the takeaway is to invest in pedagogy that AI cannot simulate - structured group work, mentorship, and feedback loops that depend on human presence.
Your membership also unlocks:
