Artificial intelligence is evolving at an extraordinary pace. Today’s AI systems are no longer static tools. They can learn, improve and be customised rapidly, making it easier and cheaper than ever for universities to develop applications tailored to their own students, educators and institutional priorities.
That shift has fundamentally changed the higher education technology landscape. What once required years of development and significant investment can now often be delivered on demand.
But while the barriers to deploying AI have fallen, the challenge of deploying it well has become even more important.
Across the higher education sector, institutions are racing to introduce new AI-powered platforms, driven in part by concerns about being left behind. The pressure to demonstrate innovation, improve the student experience and support academic staff has seen AI adoption accelerate at remarkable speed.
The results, however, have been mixed.
As universities experiment with AI, questions around governance, ethics and quality assurance are becoming increasingly difficult to ignore. Poorly implemented systems can reinforce cultural stereotypes, generate biased or inaccurate content, overlook Indigenous perspectives or produce material that is offensive or misleading. Even the most widely used generative AI tools can confidently present incorrect or culturally insensitive responses.
For AI specialists, these shortcomings are well understood. They know to verify outputs, question assumptions and treat AI-generated information critically. Expecting every educator or student to possess the same level of AI literacy, however, is unrealistic.
That makes responsible implementation just as important as technological capability.
Rather than viewing AI simply as a productivity tool, universities have an opportunity to use it to address one of higher education’s longstanding challenges: delivering genuinely personalised learning at scale.
When thoughtfully designed, AI has the potential to better support students who have traditionally been underserved by one-size-fits-all approaches. Indigenous students, learners from culturally and linguistically diverse backgrounds, students with disability and those requiring additional academic support could all benefit from AI systems that adapt content, explanations and learning pathways to individual needs.
Importantly, personalisation is not only about supporting struggling students. High-achieving learners can also benefit from AI that adjusts the pace, complexity and style of learning materials to extend their capabilities.
The technology also presents opportunities well beyond student support.
AI could assist universities in reviewing curriculum and assessment materials to identify culturally inappropriate language, unconscious bias or content that may not reflect contemporary understandings of Indigenous knowledge. Tasks that once required extensive manual review by specialist experts could increasingly be supported by AI, allowing human expertise to focus on validation and improvement rather than detection alone.
Similarly, AI can help educators and students co-design learning experiences by analysing feedback, identifying common themes and, where appropriate, creating personalised learning pathways without compromising academic standards.
Perhaps most significantly, AI offers the possibility of embedding diversity, equity and inclusion considerations into curriculum design while maintaining discipline-specific learning outcomes. Rather than treating inclusion as an afterthought, AI could help make it part of the design process from the outset.
None of these applications require futuristic technology. Many already exist in various forms around the world, while advances in AI development mean institutions can build customised solutions more quickly and at a fraction of the cost that would have been expected only a few years ago.
The question facing universities is no longer whether AI should be adopted. That debate has largely passed.
The real challenge is ensuring AI is deployed with purpose rather than urgency. The institutions that gain the greatest advantage are unlikely to be those implementing the most AI tools, but those using the technology to create more inclusive, personalised and academically robust learning environments.
