AI Acceptance and Digital Transformation as Predictive Drivers of University Sustainability Using a Hybrid Machine Learning Model

In the modern world, universities are more than just places for learning; they are complex organizations that must survive and thrive in a rapidly changing environment. To do this, they rely on two powerful forces: the shift toward using digital tools in every aspect of their work, known as digital transformation, and the willingness of their staff to embrace new technologies like artificial intelligence. Digital transformation is not merely about buying computers; it involves changing how an institution plans, manages its people, and connects with the outside world. Similarly, accepting artificial intelligence is not just about installing software; it is about whether the people within the organization believe the technology is useful, easy to use, and trustworthy enough to change how they work. When these two forces align, they have the potential to make a university more sustainable, meaning it can continue to fulfill its mission for the long term while creating value for its students, staff, and community. However, while these ideas are widely discussed, there has been little concrete proof of how they work together in the specific context of Southeast Asian universities, and no one has yet built a reliable way to predict how well a specific university will do based on these factors.

A team of researchers from Mahasarakham University in Thailand set out to fill this gap by looking directly at the people who run these institutions. They surveyed 178 high-level university executives, specifically those in charge of administration and information technology, asking them about their institution's digital progress, their staff's attitude toward artificial intelligence, and the overall health and sustainability of their university. The researchers wanted to see if the way a university transforms digitally and the degree to which its leaders accept artificial intelligence could actually predict how sustainable that university would be. They approached this question in two distinct ways. First, they used standard statistical methods to see if there was a direct link between these factors. Second, and more innovatively, they built a sophisticated computer model designed to learn from this small group of executives to predict the sustainability of any university with high accuracy.

The initial analysis confirmed what many leaders might suspect but few have proven with hard data: digital transformation and the acceptance of artificial intelligence are indeed powerful drivers of sustainability. The study found that when a university improves its digital strategy, manages its processes better, and most importantly, develops the skills of its personnel, it becomes significantly more sustainable. Among all the factors examined, the skills and digital literacy of the staff emerged as the single strongest predictor of success, followed closely by the strength of the university's external networks and partnerships. Similarly, the willingness of the university community to accept and use artificial intelligence was found to have a strong, positive effect on the institution's long-term viability. This suggests that buying technology is not enough; the people using it must believe in it and be capable of using it effectively.

To move beyond simple correlations and create a tool that could actually forecast outcomes, the researchers developed a new type of computer model. This model was designed to handle the challenge of having a relatively small number of survey responses, a common problem in studies of university leadership. The researchers started with 29 different pieces of information collected from the survey, ranging from the size of the university to specific attitudes toward technology. Using a method that mimics the way metals are cooled to remove impurities, the computer systematically tested combinations of these factors to find the most important ones. It reduced the list down to just 12 key features, cutting out the noise and focusing on what truly mattered. These 12 features included not only the digital and AI factors but also basic details about the university, such as its size and the academic rank of the respondents.

The researchers then combined three different types of learning algorithms into a single, unified system to make their predictions. One part of the system looked for complex patterns in the data, another acted like a vast forest of decision trees, and a third used a neural network to find subtle connections. These three parts worked together, with a final layer that learned how to best combine their individual insights into a single, highly accurate prediction. This hybrid approach allowed the model to learn from the small dataset without getting confused or making wild guesses. When the researchers tested this new model against older, simpler methods, it performed far better. It was able to explain more than 95 percent of the variations in how sustainable the universities were, a level of accuracy that the simpler models could not reach. The model also proved to be very stable, meaning it gave consistent results even when the data was slightly different, which is crucial for making reliable predictions.

The findings offer a clear path forward for university leaders and policymakers in Thailand and the wider region. The study indicates that the most effective way to ensure a university's future is not just to invest in new technology, but to invest in the people who use it. Developing the digital skills of staff and fostering a culture where artificial intelligence is trusted and welcomed are the most critical steps toward sustainability. The researchers also provided a practical tool for monitoring this progress. Their new predictive model can be used by educational authorities to assess the health of different institutions, identifying those that might be at risk of falling behind in their digital journey. This allows for targeted support and intervention before problems become severe.

However, the researchers are careful to note that their work has limits. Because the data was collected at a single point in time, the study shows a strong connection between these factors but cannot prove that one causes the other in a strict sense. The model was built on a specific group of 178 executives in Thailand, so while the results are highly promising, they may need to be tested with larger groups or in different countries to see if they hold true everywhere. Despite these limitations, the study provides a rare and valuable glimpse into how digital change and human acceptance work together to shape the future of higher education. It moves the conversation from abstract ideas about the digital age to concrete, measurable evidence, showing that the path to a sustainable university is paved with skilled people and a willingness to embrace the tools of tomorrow.