By Nicole Engelbert, Vice President, Oracle Higher Education Cloud
EDUCAUSE recently published its 2020 Horizon Report, and Oracle vice president Nicole Engelbert contributed her technology expertise to the publication. Here, we republish her advice to the higher education sector about artificial intelligence and machine learning.
Higher education is in a period of extraordinary exuberance around the transformative potential of technology. Vendors have invested heavily in new solutions to address admissions, instruction, student success, alumni engagement, and more. Unbridled enthusiasm for technology is not unprecedented, with institutions having navigated previous boom-and-bust cycles. The unique element of this one, however, is the largely untested belief that the success of artificial intelligence (AI) and machine learning (ML) in transforming sectors such as retail and media will transfer easily, with similar result, to the higher education industry. How many times has delivering an Amazon- or Netflix-like recommendation experience been cited as the business drive for a new technology project?
As the competitive landscape for AI and ML solutions swells, the conversation about the potential of these technologies to improve academic outcomes will become increasingly cacophonous, making it difficult for institutions to discern hype from reality and ultimately decide which solutions to select.
At least initially, the adoption paths for AI and ML will have more in common with the unruly cow paths that created London’s streetscape than New York’s planned grid, as there are myriad buyers, users, and entry points for these solutions. The reality that AI and ML will enter higher education through both intentional and unintentional choices will further intensify this disorder. AI and ML will cross the campus gates through purchases by departments, individual faculty, and even students from consumer-market sources, which might hamper these technologies’ ability to improve academic outcomes in meaningful ways, at scale.
Intentional institutional investments in AI and ML will grow through stand-alone intelligence solutions for student success, as well as enterprise systems, such as learning management systems (LMS), student information systems (SIS), and constituent relationship management (CRM) solutions with embedded or bolted-on AI and ML capabilities. Consequently, the rewards will be great for vendors able to bring rich AI and ML functionality to market rapidly, resulting in unprecedented R&D for some, marketing spend for others, and both for a lucky few. In such an environment, the rise of unclear jargon, competing claims, and vaporware is nearly unavoidable. Moreover, because the veracity of many vendors’ claims that their solutions support learning with AI and ML will be difficult to discern, establishing best practice will be painful. Taken together, these market conditions will foster uncertainty and ultimately slow institutional adoption and its impact on academic outcomes over the medium term.
On the other end of the intentionality spectrum, AI and ML will hitch an incognito ride through the campus gates on faculty and students’ consumer-market experiences. Netflix recommends TV shows for me to watch based on my profile and viewing habits, so why doesn’t the SIS recommend which courses I should take, the LMS recommend which materials I should study, or the retention dashboard recommend which students need advisement? The pressure from end users will intensify for these types of capabilities and, if unmet, will likely drive the same behavior that resulted in an explosion of feral IT over the past decade. To be certain, “the cat is out of the bag” for leveraging AI and ML for recommending services in the consumer market. It is imperative for solution providers to invest in vehicles—such as centers of excellence, independent research, and consortia—to help develop the capabilities of colleges and universities to avoid the potential implications of bias at scale or of threats to data privacy that could result from bringing these capabilities into the institutional environment.
The extent to which higher education realizes the tremendous potential of AI and ML to improve academic outcomes depends, at least in part, on technology vendors reducing the market uncertainty surrounding new solutions. Progress can be made in three key ways:
“With great power comes great responsibility.” Whether attributed to Voltaire or Stan Lee, this quote is deeply relevant to the rise of AI and ML technologies in higher education. Technology vendors must be better partners. We must see ourselves as members of the higher education community, actively fostering the kind of industry that advances our society. A commitment to education, transparency, and social responsibility will be critical to ensuring the best future with AI and ML. The genie is out of the proverbial bottle; now we must help ensure that it is a force for good.
Nicole Engelbert is responsible for engaging institutions globally to inform the development of Oracle’s Student Cloud solution. Prior to Oracle, she served as the Director of Research & Analysis at Ovum, where she advised institutions on their technology strategies. Engelbert holds a BA in classics from Union College and an MEd in educational administration and policy analysis from Columbia University.