The outcome could be just as psychological as the economy, as universities are not preparing students to participate in the AI-transformed labor market, according to a leading economist.
Tyler Cowen, a professor at George Mason University, said the university “is making a generation of students who are out on the labour market and are completely unprepared for what they are expected of.”
Students are being taught “even counterproductive” and “even counterproductive” in a conversation with podcaster Azem Azhar aired Tuesday.
He said studying in college is “time to do that,” and refers to learning to work with AI.
Although Cowen did not name any particular discipline, his comments come in the discussion of how AI can reconstruct the type of knowledge work that college students expect after graduation. As generative AI tools become more integrated into the white-collar industry, proficiency in these systems is becoming a baseline expectation.
Cowen said he doesn't think AI will destroy jobs, but he will restructure employment practices, career trajectories and productivity standards.
Cowen said new graduates could struggle to get hired, warning that the damage could be deeper than wage losses.
“Even though production is low, I think a lot of the best costs are psychological. People feel they don't fit into this world,” he said. “And they're probably a bit right.”
The destructive power of education
Cowen's warnings join a growing chorus of concern that elite institutions and traditional curricula have not changed quickly enough to keep up with rapid technological changes.
Last week, Leah Belsky, Vice President of Education at Open AI, said in a company's podcast that university graduates should know how to use AI in their daily lives and should teach them to use “in a way that expands critical thinking and expands creativity.”
Google Deepmind Research Scientist Stefania Druga told Business Insider in May that if AI can complete a student assignment, teachers will need to change it.
“If AI can solve a test, it's the wrong test,” she said.
Draga, which helps design AI education platforms for children, said that instead of being taught young people to use it as a tool for co-creation, they are using the full learning.
Educators employ divergent strategies. Some have doubled analog tools such as handwritten essays and oral exams to maintain academic integrity, while others have embraced AI and built more personalized, game-hard reviews.

