The future of digital learning in the age of AI

Admitting it might be a bit uncomfortable, but many decisions in digital learning content are still based on theory and intuition. Whilst theory is valuable, some of it remains unproven. Intuition, on the other hand, often carries biases that we can all acknowledge. So, what if there was a different way to truly understand how people learn through digital experiences? Could machine learning help us reverse engineer the learning process?

AI presents an exciting opportunity for disruption and innovation in digital learning. These advanced systems can analyse vast data sets to create educational experiences that are not only personalised but also backed by empirical evidence. This approach moves beyond traditional methods that often rely on subjective ideas about interactivity and engagement, which are sometimes added based on a designer’s gut feeling.

Take, for example, the common drag-and-drop exercises in online courses. Traditionally, these were included under the assumption they boost engagement. However, with AI-driven insights, such decisions can now be based on data that proves their effectiveness in specific contexts. This shift towards evidence-based design ensures that educational content is both engaging and effective. This isn’t to say everything we’ve done so far is flawed or incorrect! We’ve worked with the tools and knowledge available to us. But the future offers a different landscape.

Despite being an improvement over traditional learning methods, personalised learning has faced challenges with biases — often stemming from a designer’s preferences or pressure from subject matter experts (who might be footing the bill). If you disagree, just think back to Learning Styles. AI can help mitigate these biases by analysing a diverse range of user experiences, fostering more equitable learning environments.

Machine learning models excel at recognising patterns across large datasets, allowing them to make predictions and take actions accordingly. This capability is essential for creating genuinely personalised and effective learning experiences for a wide array of learners.

The shift to data-based learning design marks a fundamental change in how we create and deliver learning content. By leveraging AI to analyse past user interactions and outcomes across thousands of users, we can design courses more likely to succeed in meeting their objectives. This approach not only improves educational quality but also ensures learners receive the most relevant and impactful content.

In practice, this means every aspect of a course — from its structure to its interactive elements—will be scrutinised through data analytics. This doesn’t mean the art of designing learning experiences becomes automated; rather, it adds scientific rigour to the craft. We’ll gain insights into what works and what doesn’t, enabling us to make informed decisions about course design.

AI can play a crucial role in assessing learner progress, providing real-time feedback, and even predicting future learning needs. By harnessing these capabilities, we can create dynamic and responsive learning environments that adapt to individual learners’ needs. This would happen through AI agents—a new frontier for many learning professionals—and as we look towards this future, I’m confident that AI agents will become integral components of learning ecosystems, potentially even replacing the need for traditional courses.

But what might a future without formal courses look like? Perhaps it involves interacting with a digital agent that triangulates data from multiple aspects of your work and life to create bespoke learning experiences. Imagine a language course tailored specifically to your circumstances—like having a German mother-in-law (mine is absolutely lovely, danke)—or an agent that notices your upcoming flight to Frankfurt and focuses on travel-related phrases. The impact could be profound.

This vision isn’t just about embracing new technology for its own sake; it’s about enhancing the quality and accessibility of education for all learners. By focusing on data-driven approaches, we can ensure educational experiences are both effective and equitable.

Join me at the OEB Global Conference as we explore these topics further and chart a course for the future of digital learning in the age of AI.

Written for OEB Global 2024 by Lori Niles-Hofmann. Join Lori in Berlin at her session, ‘Data-Driven Learning Design in the Age of AI‘, on Friday, November 29 2024.