A new study finds that universities remain lagging in adjusting their curricula, assessment methods, and career preparation as AI becomes more widely adopted in businesses and the public sector. The study argues that institutions should move beyond focusing on detecting AI-generated content and academic misconduct, and instead directly address evolving workplace demands for student skills.
Research focuses on the shift toward teaching
The study, published in Frontiers in Education and authored by Kelechi Ekuma of the Institute for Development Policy at the University of Manchester, states that since ChatGPT was launched to the public in 2022, many universities have primarily responded to generative AI by focusing on detection tools and plagiarism policies—but these measures do not address how students will collaborate with AI in the workplace.
Research indicates that AI and automation have entered fields such as public administration, welfare distribution, agriculture, finance, healthcare, education, and workforce management. This means that universities are no longer just facing new tools entering the classroom, but rather fundamental changes in knowledge production, teaching methods, and professional environments.
Critical AI literacy has been highlighted.
The author argues that universities should make "critical AI literacy" a key focus of their curricula. At its core, this goes beyond merely using tools—it includes understanding how AI works, where it may fail, and how to make judgments, evaluate ethical implications, communicate clearly, and adapt to new technologies in complex situations.
Research suggests that viewing AI primarily as an academic integrity issue may cause schools to overlook longer-term capacity building. The article also highlights risks associated with AI applications, including errors, bias, overreliance, unequal access, and the influence of major tech companies on system development and distribution.
The government and schools have already been advancing training.
In this context, the author recommends that universities allocate more resources to developing skills that AI cannot easily replicate, such as critical thinking, ethical judgment, communication, and understanding complex social issues. The study also emphasizes that this does not mean every course must be transformed into an AI course, but rather that existing courses need to be reevaluated to consider how AI will change the problems and methods traditionally taught.
This study is being released as schools, businesses, and government agencies accelerate their AI training initiatives. The U.S. Department of Labor has launched an AI apprenticeship portal to expand training coverage in fields such as education, finance, healthcare, and manufacturing.
Earlier this year, Google’s charitable arm announced a partnership with the Sundance Institute to launch a $2 million initiative aimed at training over 100,000 artists in using AI tools. In April, U.S. President Trump signed an executive order establishing a White House AI Education Task Force and directed federal agencies to expand AI programs for students and educators. That same month, Mississippi College School of Law began requiring first-year students to complete an AI-related course focusing on understanding the technology and verifying its outputs.