AI at Work in Higher Education: What Practitioners Are Doing, What They’re Worried About, and What It Means for Institutional Readiness

Most respondents (81%) feel enthusiasm or a mix of caution and enthusiasm toward AI. Just 17% said they feel cautious or very cautious. 

Most respondents' institutions (92%) have a work-related AI strategy. The most common elements of those strategies are piloting AI tools, evaluating both opportunities and risks, encouraging staff and faculty to use AI tools, and creating policies and guidelines. 

Among respondents who said their institution's work-related AI strategy includes increasing workforce skills, a majority (69%) said they are doing so primarily by upskilling or reskilling existing staff and faculty. 

Just 13% of respondents said their institution is measuring the return on investment (ROI) for work-related AI tools, pointing to a significant opportunity for institutions to embark on such work. 

Though nearly all respondents (94%) said they have used AI tools for work within the past six months, only 54% of respondents said that they are aware of policies and guidelines meant to guide such use. This gap could have serious consequences for data privacy and cybersecurity, data-informed decision making, digital accessibility, and more. 

More than two-thirds of respondents (67%) identified six or more risks as "urgent," indicating that most respondents are concerned about a breadth of risks associated with using AI. The three most-selected risks are an increase in misinformation, the use of data without consent, and a loss of fundamental skills requiring independent thought. 

Similarly, more than two-thirds of respondents (67%) identified five or more opportunities as "most promising," indicating that most respondents are also optimistic about a range of opportunities for using AI. The three most-selected opportunities are automating repetitive processes, offloading administrative burdens and mundane tasks, and analyzing large datasets. 

The three most-selected challenges associated with using AI tools for work are AI's pace of change, lack of AI expertise, and lack of best practices. 

Among respondents who have recently used AI tools for work-related tasks, a majority (73%) reported that they do so either daily or weekly. 

Among respondents who have not recently used AI tools for work-related tasks, the most common reasons they gave were that they have not wanted to (58%) or have not needed to (50%). 

Just 11% of respondents reported that they are required to use AI tools for work, while 64% reported that not only are they not required to use AI tools for work but they do not expect to be required to do so in the near future 

A majority of all respondents (86%) indicated that they want to use or continue to use AI tools in the future. 

From a list of 30 potential work-related uses of AI tools, more than half of respondents (54%) said they had used AI tools for eight or more types of tasks in the past six months. 

Most respondents have at least some access to AI tools for work through their institution. Just 10% indicated that their institution does not provide access to any of the AI tools they want to use for work. 

More than half of respondents (56%) reported that for work-related tasks, they have used AI tools that are not provided by their institutions. Similar to findings about lack of awareness of policies and guidelines, this insight points to another area of AI-related risk for institutions.