North-West University (NWU) has adopted an official AI policy to provide a framework for the safe and secure management and governance of AI within the university, says NWU AI Hub director Professor Anné Verhoef.

The framework was needed because the university's AI strategy focuses on embracing AI in a human-centred, ethical and responsible way, and it also helps shape national conversations about the responsible use of AI in higher education and research, he adds.

The policy encompasses teaching and learning, assessments, research and postgraduate supervision, all of which need to be coordinated, supported and managed within a clear framework.

“NWU's AI policy delineates the roles and responsibilities of students, lecturers, researchers and management. Universities across the world are moving in a similar direction to ensure a human-centred approach to AI,” he says.

Further, the university’s AI policy clarifies these roles and establishes an effective governance model, says Verhoef.

“Many universities in Africa and globally are encountering challenges in creating a comprehensive and centralised approach to the coordination and management of AI. Without a harmonised strategy that integrates teaching, learning, research and management, universities face conflicting approaches that can confuse and disadvantage students and lecturers,” he says.

The university's AI policy resolves these challenges by clearly defining stakeholder roles and locating AI governance within the IT department, and recognising that AI is fundamentally an IT tool or application.

“The AI Steering Committee, which includes representatives from students, faculty, administration, the library and the Centre for Teaching and Learning, is responsible for all AI-related matters at the university. This committee reports to the IT Committee, which in turn reports to the University Management Committee. Communication then flows to faculties and to Senate as required.”

Additionally, the university's council undertook an extensive consultation process involving faculties, Senate and refinement committees before it approved the policy.

“The NWU’s AI policy was developed through a comprehensive consultation process, and this collaborative approach ensured that the policy addressed a wide range of expectations and challenges from all stakeholders,” Verhoef says.

Further, many of the concerns raised were philosophical rather than technical, he notes.

Maintaining a human-centred approach to AI was a major focus, rather than only identifying which AI tools to use. The policy also emphasised the need to address AI’s significant energy and water consumption, as well as its ecological and environmental impact, which highlighted the importance of sustainable AI practices, he adds.

With this policy, the NWU aims to set a clear national benchmark for responsible, human-centred and sustainable AI in higher education.