The Causal AI/ML Revolution in Education

Artificial Intelligence (AI) and Machine Learning (ML) have become ubiquitous in our everyday lives from consumer applications to enterprise systems. Predictive analytics, a field of machine learning, has gone from a nascent concept over ten years ago in student success to now a critical component to improving outcomes in all areas of education.

While predictive analytics has “grown-up”, there still remains questions and concerns about its use in education. Specifically, concerns around black-box algorithms, trust in prediction scores, using past data to model the future in an ever changing world of pandemics and demographic shifts in Higher Education. As such, the activity of solely using machine learning to train and test models puts into many questions its ability to truly shed a light on the right key drivers for student success.

Today, however, we are witnessing a new development in AI / ML that has been around for nearly two decades but is just now emerging as a new tool to help education achieve the outcomes it has been searching for. It’s called Causal AI / ML and it may offer us a major advance in understanding the cause and effect of student success initiatives and the efficacy of edtech investments, and do it affordably.

While Causal AI / ML has quietly been used in scientific pursuits for years, it's no secret to advanced start-ups in the world including the likes of Google, Lyft, Netflix, and Uber. It is emerging as a proven analytics approach that can be used at scale. Most importantly, it is now being used to improve student success.

And unlike predictive analytics, which imparts a sense of finality or inevitability, Causal AI / ML helps students and educators improve their lives and prevent negative outcomes through interventions and feedback that produce a high return on investment. Institutions or businesses that are adept at using evidence from both retrospective and prospective data are gaining an advantage.