Generative artificial intelligence (AI) is already part of learning, teaching, and administrative processes at universities. An analysis published by UNESCO reports that about 86% of students worldwide use these tools regularly in their studies and estimates that nine out of 10 faculty members and researchers rely on them in their professional work.
However, even though this technology is being adopted at a rapid pace, many higher education institutions still lack clear rules for using it in an ethical and responsible way.
Against this backdrop, researchers from IFE Europe at the Institute for the Future of Education of Tecnológico de Monterrey are participating in Integrity AI, an international research and educational innovation project designed to help universities prepare to govern the use of AI.
“There is a gap between the pace of technological adoption and the level of institutional readiness,” explains Virginia Rodés, a researcher at IFE Europe. “Most institutions still lack governance systems, training programs, and legal and regulatory frameworks to ensure the sustainable, responsible, and sovereign use of AI.”
The project is led by the Institute for the Future of Education (IFE Europe) and Mondragon Unibertsitatea. It is funded by the European Union’s Erasmus+ program through the Capacity Building in Higher Education initiative and brings together universities from Latin America and Europe to develop governance frameworks, including policies, tools, and strategies for the use of AI in higher education settings, while strengthening academic integrity.
AI in Higher Education: Universities Are Struggling to Regulate It
In the project’s first phase, the team conducted an assessment to evaluate how prepared higher education institutions are for AI. Through workshops and surveys involving 148 participants from seven universities across Latin America, including faculty members, researchers, administrators, administrative staff, and students, the researchers examined how these tools are being used and how ready campuses are to govern them.
The assessment included Mexico’s Instituto Politécnico Nacional and Universidad Autónoma de Nuevo León; Colombia’s Universidad de San Buenaventura de Cali and Universidad del Cauca; Guatemala’s Universidad de San Carlos and Universidad Galileo; and Brazil’s Universidade Federal de Santa Catarina.
Fifty-five percent of participants said they use AI frequently or very frequently for teaching, learning, research, and administrative tasks. The study also found that 57% of respondents reported a moderate level of familiarity with these tools.
The findings reveal that AI has already become part of the day-to-day activities at these universities. Among the benefits identified were time savings when completing tasks, increased productivity, assistance with writing and translation, and improved access to information.
The study also identified challenges, including inaccuracies and biases in generative AI, inadequate source verification, students’ cognitive overreliance on these tools, and difficulties related to citation, content authorship, and data privacy.
However, the most significant challenge identified was the lack of governance: every institution reported either the absence of or insufficient regulatory frameworks to oversee the use of this technology.
“The gap does not lie in awareness, but in how that awareness is translated into operational practice,” she says. “University communities recognize the risks and ethical dimensions of using AI, but they lack operational policies, shared criteria, protocols, and accountability mechanisms.”
Setting Clear Rules
In its next phase, the project will focus on developing governance models tailored to the needs of the participating universities. The goal is to design institutional policies, training programs, assessment tools, and AI integration mechanisms that are also aligned with each institution’s values.
The initiative will also move into a pilot implementation phase at the participating institutions, where the governance models developed through the project will be tested. This stage will include the creation of specialized AI units and the development of guidelines covering areas such as transparency, AI disclosure and use, human oversight, institutional ethics review, and the protection of personal data.
At the start of the pilot programs, all participating universities will implement a shared set of core actions. These include formally establishing an AI governance structure, developing an initial ethical and institutional framework, creating a dedicated channel to address questions about the use of AI, and launching training and awareness activities for their university communities.
Building on that foundation, each institution will define its own strategy and may introduce additional measures based on its specific context and needs.
Looking ahead, the team hopes the model developed through Integrity AI can be adapted and replicated at other universities across Latin America and Europe. To support that goal, the project will produce implementation guides for ethical and institutional frameworks for the responsible adoption of AI, along with training programs designed for different members of the university community.
“It is a collaborative effort to develop best practices within some institutions while also placing a strong emphasis on creating an impact beyond them by developing tools that other universities across the region can adopt,” Rodés says.
Dominique Boullier, a sociologist and professor emeritus at the Paris Institute of Political Studies (Sciences Po), says initiatives like this allow universities to establish their own criteria for evaluating and adopting AI, rather than simply keeping pace with the agenda set by the technology companies developing these tools.
He argues that universities should promote coordinated evaluation efforts across institutions that assess not only the performance of AI systems but also how they are integrated into different educational settings and the value they truly add to learning.
“If we move toward an internationally coordinated process for testing every specific use of these systems, not only their capabilities, but also the way they are embedded within educational environments, we can compare results and bring all that knowledge together so that everyone can draw on those lessons in their own contexts,” Boullier says.
Were you interested in this story? Do you want to publish it? Contact our content editor to learn more marianaleonm@tec.mx
