The global higher education community is in the midst of irreversible changes to the learning experience brought on by the pandemic. Blended classrooms and hybrid learning are here to stay. But in the fallout of the pandemic, longer-term consequences pose a serious threat.
Recent McKinsey research examining higher education finds that the risk of “outcome inequities” for students could worsen in the U.S., impacting completion, employment, and lifetime earnings. This, even as colleges battle falling revenue and declining enrollments with a business model at its “breaking point.” Universities that quickly ramped up digital infrastructure early in the pandemic now need a more measured, long-term digital transformation roadmap to help address the challenges brought on by COVID-19.
Any such roadmap should include artificial intelligence (AI), which has the potential to revolutionize higher education, with advancements offering enormous potential to tackle hard problems that intensified during the pandemic. But knowing the value of AI and adopting AI are two very different things—and for most universities, the challenge of bridging that gap remains.
Higher ed stakeholders are generally reluctant to adopt new technologies before they have been sufficiently de-risked elsewhere, and are especially reluctant to do so when resources are limited. It’s one thing for AI to recommend the next item on your online shopping cart. It’s another for technology to recommend an online curriculum that can directly impact a student’s ability to find or do a job.
What’s clear is the institutions that succeed in integrating AI into their transformation roadmap will have a distinct advantage over their competitors, which will translate into higher enrollments, higher revenues, and higher rates of student achievement.
So for all those universities sitting on the sidelines but eager to join the fray, here are three strategies to get you started down the right path to AI-assisted education:
1. Identify the highest-priority problems you’re trying to solve.
The best way to facilitate AI adoption is to begin by asking, “What problems are we trying to solve?” In a worrying trend, a recent Gallup survey found that one-third of students considered withdrawing from courses in the past six months, citing “COVID-19 and emotional stress.” More than half reported that COVID-19 would likely impact their ability to complete their degree. Supporting students at risk is critical, with AI providing the technological heft to scale personalized, high-touch support. Through its machine learning solutions, Copenhagen-based Damvad Analytics has helped universities identify 80 percent of students at risk of dropping out more than six months in advance, allowing institutions to intervene early.
Naming the challenge may sound simple, but it’s a step that’s often overlooked. In the corporate world, more than $30 billion is wasted on unused technology each year by companies that purchase unnecessary solutions for low-priority problems. Employees end up wasting time and failing to adopt these new technologies in any meaningful capacity.
The same holds true for universities and other organizations looking to integrate technology into their everyday processes. Once identifying a problem or set of high-priority problems, it is easier to discover how AI technologies can be applied to provide an effective solution.
2. Build consensus around new and emerging technologies, and leverage the experience of early adopters.
Amid COVID-19, hybrid learning environments have become the new normal. And while this grew out of a public health necessity, higher education institutions have begun to realize the benefit of hybrid and online learning, especially as a means to reach and teach more students. AI can help professors build efficiencies when teaching at scale. Duke University uses AI-assisted grading tool GradeScope to help streamline grading and provide more consistent feedback to students. The tool was particularly relevant for instructors looking to conduct assessments remotely during campus shutdowns.
But while moving in this direction of “scalable education,” the largest obstacle to adoption is usually the university’s comfort level embracing new technologies. Do stakeholders trust the tools themselves? Do teachers and administrators trust their ability to put these technologies to good use? Are there faculty members who have ethical or bias concerns with AI being integrated into their classrooms? These questions must be addressed first, as they will determine the success of AI integration into the classroom.
In order to get over these adoption hurdles, it’s important to seek out early adopters and to emphasize the ways in which technology enables faculty members and stakeholders to be more effective and spend their time on activities they feel are crucial and important–while AI handles the busy but repetitive tasks.
3. Start with one pain point—and then crawl, walk, run.
Once you have pinpointed an unmet need within the organization and achieved collective buy-in from necessary stakeholders, the next step is to implement the appropriate application of the technology, gather some data, reflect on the results, and scale the implementation.
For example, if your top pain point is around graduation and retention rates for students, implement a solution and then track the progress of batches of students before rolling it out across the entire organization. AI and machine learning tools will also be able to point out where you can continue to improve. You may find that retention issues happen most often within a certain lesson, or student engagement falls dramatically right after difficult chapters. With this type of real-time information, you’ll know where to focus your efforts to improve learning outcomes for students.
Before COVID, the thought of integrating technology this intimately in the learning process felt intimidating. Even public conversations about the role AI might play in higher education feared automation over personalization, metrics over meaningful relationships. Educators wondered if this was the beginning of the end for their jobs. That’s an unlikely scenario. A Microsoft-THE survey found that university leaders do not anticipate cutting teaching staff over the next 10 to 15 years as a result of integrating AI. If anything, 25 percent expect to recruit more teaching staff–a consequence of more students seeking out a quality education to remain employable, with AI and automation on the rise.
As we’ve seen throughout 2020, technology and higher ed are not opposing forces. More than ever, educators and learners value the opportunities unlocked at the intersection of tech and education. Such collaborative experiences are going to reshape the future of higher education.
- Generative AI can enhance equity of access and attainment - December 7, 2023
- The academic implications of AI in student writing - December 5, 2023
- New program aims to train more male educators of color - December 4, 2023