Most Campus Tech Leaders Say Higher Ed Is Unprepared for AI’s Rise

This post was originally published on this site.

Nearly two years after generative artificial intelligence exploded into the public consciousness, just 9 percent of chief technology officers believe higher education is prepared to handle the new technology’s rise.

That’s one of several key takeaways from Inside Higher Ed’s third annual Survey of Campus Chief Technology/Information Officers, executed with help from Hanover Research earlier this year. The results released Wednesday offer a portrait of how colleges and universities are—and aren’t—using technology to enhance learning, work and research in an era of rapidly evolving digital innovation. 

About two out of three CTOs said the digital transformation of their institution is essential (23 percent) or a high priority (39 percent). And most are concerned about AI’s growing impact on higher education, with 60 percent worried to some degree about the risk generative AI poses to academic integrity, specifically.

Despite ongoing worry from administrators and educators that AI will increase cheating, 46 percent of CTOs are either extremely or very enthusiastic about AI’s potential to boost their institution’s capabilities, according to the report.

But that doesn’t necessarily translate to institutional enthusiasm for prioritizing digital transformation or investing in AI technology. CTOs believe other senior leaders at their institution are less likely than they are to view digital transformation as a top priority. And not quite one in four said investing in artificial intelligence is an essential (1 percent) or high (22 percent) priority for their institution.

Still, this is an increase from Inside Higher Ed’s 2023 CTO survey, when just 16 percent of CTOs described investing in AI as a top priority for their institution.

“People are warming up to the idea,” said Ravi Pendse, chief information officer at the University of Michigan, “but there’s also some people who are passionately opposed to it, and that’s OK.”

He likened the adoption of AI to the rise of smartphones in the early 2010s: While some people still use flip phones in 2024, the majority of people now use a smartphone because its capabilities—such as rideshare and banking apps—make daily tasks easier and more efficient.

“The great thing about a university culture is that no one is going to force you to use [AI], but students are certainly using it, so institutions may as well get onboard,” Pendse said.

Enterprise-Level AI Approaches Lag

But the adoption of AI technology varies in scale and scope.

Almost half of CTOs who responded to the survey agreed that their institution places more emphasis on thinking about AI for individual use cases than on thinking about AI at an enterprise level. Fifty percent of CTOs said their institution is using it to create virtual chat bots and assistants, which was the most popular application.

That aligns with what presidents, provosts and chief business officers reported in their own Inside Higher Ed surveys, which also showed that chat bots and virtual assistants are the most common use of AI by colleges and universities.

The CTO survey found that the most popular applications of AI after chat bots included enhancements of cybersecurity (35 percent), predictive analytics for student performance and trends (27 percent), research and data analysis (24 percent), and learning management systems (20 percent).

Only 3 percent of institutions use AI for facilities management or creating personalized learning pathways; 11 percent use it for institutional planning and decision-making, student engagement, and fundraising; 17 percent use it for admissions and administration processes (such as scheduling and resource allocation); 14 percent use it for student advising and support; and 12 percent use it for grading and assessment.

Some 18 percent of CTOs said their institution does not use AI for any of those purposes.

“Each institution has different levels of resources, both financial and technical. When these new technologies come up, it’s much easier to implement a small cluster here and a small cluster there,” Pendse said, noting that the University of Michigan is one of a handful of institutions that has already integrated AI into numerous universitywide functions, including those related to teaching, learning and research.

Those efforts were guided by a report produced by the university’s generative AI committee in 2023, which aimed “to set the stage for the [University of Michigan] community and society at large to harness the transformative benefits of GenAI, while simultaneously mitigating its inherent risks.”

Pendse, who led the committee alongside Michigan’s provost, said buy-in from university leadership is critical to scaling AI technology. “Once they’re aligned, you can have enterprise-level conversations,” he said.

While most CTOs somewhat (38 percent) or strongly (50 percent) agree that their central IT department has found effective ways to reach out and partner with other areas of the institution, more than half of CTOs agree, somewhat (37 percent) or strongly (20 percent), that senior administrators at their institution treat the central IT unit more like a utility than a strategic partner.

What’s more, most institutions don’t have the same resources as Michigan, which had a $17.9 billion endowment in 2023, to explore how AI technology can work to their larger, long-term benefit.

Nonetheless, 61 percent of CTOs said their institution experienced no central information technology budget cutbacks in 2023–24. Most CTOs expect their 2024–25 central IT budget to be about the same as last year’s, if not higher, with about half of CTOs reporting their institution’s annual IT operations and services budget to be between $1 million and $5 million.

Still, that may not be enough for colleges and universities to adopt large-scale AI approaches.

AI is still a relatively emerging or developing capability for institutions, and many of us are still trying to figure out what the best use cases and biggest risks are with using these technologies.”

—Mark McCormack, senior director of research and insights at Educause

According to a recent AI landscape study from Educause, the professional association for higher ed IT professionals, few institutions had AI-related acceptable use policies in place, and even fewer institutions had implemented workforce changes to accommodate the staffing needed to support enterprisewide capabilities.

“AI is still a relatively emerging or developing capability for institutions, and many of us are still trying to figure out what the best use cases and biggest risks are with using these technologies,” Mark McCormack, senior director of research and insights at Educause, wrote in an email. “Institutions will need to build up their policies and institution-wide governance, strengthen their infrastructure (including required staff, funds, and technologies), and improve staff, faculty, and student AI literacy as important next steps in evolving and maturing in this area.”

The findings of Inside Higher Ed’s CTO survey support that idea, showing that more than half of CTOs say their institution hasn’t adopted a policy or guidelines in the areas of instruction, administrative tasks, student services and research assistance. Just 12 percent of CTOs reported that their institution has a comprehensive policy overseeing AI security and ethics, and 29 percent said they have guidelines for using AI in instruction, such as tutoring and content generation.

A plurality of CTOs (48 percent) also said their institution does not have sustainability goals related to technology use. Just 5 percent say their institution’s use of AI has increased its carbon footprint/electricity use greatly or “extremely.”

Lukewarm on Remote Work

Developing effective guidelines and implementing AI technology at scale will require workers with AI expertise. But according to the survey, 68 percent of CTOs said their institution is struggling to hire new technology employees, and 40 percent somewhat or strongly agree that their institution is struggling to retain technology employees. The vast majority of CTOs attributed recruitment and retention challenges to more competitive offers outside higher education.

While higher education institutions typically don’t have the resources to pay their employees like private tech companies would, remote and flexible work has become a nonmonetary benefit prospective employees are increasingly drawn to post-pandemic.

However, the survey shows that about half of institutions don’t have policies that encourage remote/flexible work, despite 98 percent of CTOs who agreed that their institution has the technology to make it viable and 86 percent who said their colleagues expect more flexibility to work when and where they want to now than they did before the pandemic.

That finding aligns with a 2023 survey from the College and University Professional Association for Human Resources, which found that 44 percent of nonfaculty higher education employees said that limited remote work options was one of their top reasons for looking for another job.

While student-facing roles require in-person work, the increased appetite for remote work for jobs, such as those in IT, that don’t necessarily require a physical office presence is “going to change the way higher education looks,” said Muhsinah Morris, an assistant professor of education at Morehouse College and director and principal investigator of Morehouse in the Metaverse, a virtual learning environment.

“It’s not as if higher ed is paying so much,” she said. “With remote work, you can save gas money and time on commuting … In this economy, all of those factors matter.”

Cybersecurity in the Age of AI

What also matters, Morris said, is the CTO survey’s finding that few university tech officers are very confident that their institution’s practices can prevent cyberattacks that could compromise data or intellectual property, or lead to a ransomware event. Sixty-three percent said they’re moderately confident.

“It tells us that they absolutely are not ready to take on what’s going to happen when the floodgates of AI open up and people have access to more robust techniques to attack institutional networks,” she added.

Morris noted that higher education institutions—especially high-profile and research-intensive institutions—are particularly vulnerable to cyberattacks because not only do they manage the personal information of students and staff, they also manage databases that host proprietary, patentable and trademarked research.

And at institutions that aren’t thinking about AI from a cross-campus, enterprise level, that vulnerability is even greater as malware and phishing get more sophisticated by the day.

“If they don’t have any understanding of how these cyberattacks are happening in the age of AI,” Morris said, “it makes an institution vulnerable for a hostile takeover.”

More Key CTO Survey Findings

• Regarding their institution’s priority areas for digital transformation efforts, many CTOs (81 percent) cited student success. Libraries came in last in a list of possible priority areas (26 percent). About a third of CTOs (35 percent) said their institution has set specific goals for digital transformation.
• Two in 10 CTOs (20 percent) said their institution has partnered with a technology company to implement AI. An additional three in 10 (32 percent) said their institution is considering such a partnership. And about one in seven CTOs each said their institution is currently building its own technology using open-source AI models (15 percent) or considering doing so (also 15 percent).
• College and universities continuing to lean into online instruction by various metrics, with public institutions having an apparent edge over private nonprofit ones: Some 96 percent of public institution CTOs somewhat or strongly agreed that their college or university will sustain its ability to offer high-quality online courses, compared to 37 percent of private nonprofit institution CTOs.
• Most CTOs somewhat or strongly agreed that their institution supports teaching with technology in technical ways, such as by providing technical support for teaching online courses (84 percent). Fewer CTOs agreed, somewhat or strongly, that their institution supports teaching with technology in other ways, such as by considering it in tenure and promotion decisions (39 percent) or by appropriately accommodating for the time demands of online courses on faculty workloads (33 percent).