Generative AI (GenAI) will bring innovations, efficiencies, creativity, and effectiveness to most all people who work at our colleges and universities in the coming year.

In the previous edition of Online: Trending Now, we looked at the AI tools and activities of the AI-augmented professor. Yet there are more staff members who support the learning process at most universities than there are those who directly teach the students. Let’s take a look at how AI will facilitate the work of the instructional designers, researchers, administrators, and other non-teaching professionals in colleges and universities this fall.

This comes in the context of Microsoft and LinkedIn’s 2024 Work Trend Index Report which surveyed some 31,000 people across 31 countries, uncovering rather surprising facts along the way. Bret Kinsella in Synthedia writes: “The report found that 78% of knowledge workers bring their own AI (BYOAI) to work. For GenZ, the figure is 85%. However, even Baby Boomers are BYOAI at a rate of 73%.” This phenomenal trend has unfolded in just one year. Citing the same large survey by Microsoft and LinkedIn, Sabrina Ortiz writes in ZD Net that “AI skills are so much of a priority that the report suggests 66% of business leaders wouldn’t hire someone without AI skills, and 71% of leaders would prefer to hire a less experienced candidate with AI skills than a more experienced candidate without them.”

Many instructional designers are among the BYOAI group who use GenAI tools on a daily basis to create graphics, images and audio segments for university classes. Having an AI-enhanced toolset available at home as well as work allows for seamless hybrid working that appeals to many designers. While OpenAI’s Sora that generates full-motion videos from prompts, is only available to a limited number of testers now, it is expected to be released soon. Content generation can be tailored for transitions by instructional designers, who often use multiple chatbots with citations to ensure accurate and precise, quotable materials for classes. The flow and design of classes, culminating in learning outcomes, may also be recommended by GenAI to the instructional designer.

A favorite approach of mine to learning design is mastery learning. While mastery learning modules can be constructed manually, a number of automated, AI-driven systems are available to speed up the process and bring a standardized process to the delivery and assessment of learning, while enabling personalization of learning along the way.

University researchers have been using AI applications for years. The new release of multiple expansive large language models facilitates the collection and analysis of data that can be run in a secure environment, “air-gapped” from the public internet and secured on university computers that are not connected outside the research environment. The integration of many of the current chatbots with common tools of analysis, presentation and publication facilitate the creation and delivery of research reports. Further, GenAI can uncover less obvious correlations, causes, effects and trends that might otherwise not be obvious to researchers. Especially when working with multiple datasets, new associations can be uncovered that can make the research much more valuable.

Deans, directors and department heads or chairs also have much to gain from use of GenAI. Of course, we must be careful not to use confidential personnel or financial data in the prompts that we submit to open large language models. Depending upon the level of security, volume, complexity and isolation of data that is required, there is, however, the advent of Small Language Models (SLMs) that are surprisingly powerful when operating on small platforms, such as smart phones. With connectivity turned off, these devices are not connected to the internet (air-gapped), maintaining security and privacy.

As Joinal Ahmed of Microsoft reports “Small Language Models (SLMs) represent a focused subset of artificial intelligence tailored for specific enterprise needs within Natural Language Processing (NLP). Unlike their larger counterparts like GPT-4, SLMs prioritize efficiency and precision over sheer computational power. They are trained on domain-specific datasets, enabling them to navigate industry-specific terminologies and nuances with accuracy.” This means that privacy is retained by maintaining all prompts, data, processing and results in an environment that is not connected to the cloud.

In this secure environment, administrators can ask apps to perform personnel, budget, and organizational comparisons, analyses and recommendations without exposing the data to possible discovery by others. For example, an administrator can input the college or department budget onto the app and perform an offline analysis of expenditures as well as prompt the app to run budget scenarios for 10% or 20% reductions while optimizing outcomes such as ROI, student enrollment and student success.

In less-secure outward facing analyses, an administrator can ask for a search of public records to identify details of the student enrollments in analogous departments at competing universities. GenAI can compare publicly available enrollment demographic data to construct a more effective enrollment plan and associated marketing budget. In addition, administrators may want to project future industry hiring in fields associated with degrees and certificates offered or planned by the university. The GenAI tools can find and assemble data that we simply did not consider in the past for powerful analyses and recommendations.

In the highly-competitive atmosphere of higher education recruiting, American universities are often pressured by uncertain funding and political interventions that may modify long-standing policies upon which the public had relied. In response to these pressures, universities must have access to the best data, the most powerful analyses, and the most innovative recommendations to enhance recruiting and enrollment in order to cover the ever-rising expenses. We are now facing the closure of one college or university every single week in 2024. GenAI can provide the creativity, backed by sound data and analyses that might make the difference between a balanced budget and the slippery slope toward closure.

This article was originally published in Inside Higher Ed’s Transforming Teaching & Learning blog. 

It is early August 2024.  I am about to begin the fall term of teaching, research, administrative tasks and advising with the help of Generative Artificial Intelligence (GenAI) tools and assistants. 

My syllabus was already submitted a couple of months ago so it is available online to all who may want to register. I did the annual update using a couple of GenAI chatbots that added a new lecture or two, some new graphics and a new introductory field-specific vocabulary quiz. They also created upgraded new authentic assessments to adjust to changes in the field, keeping the material up to date and relevant. The grading rubrics were refined and adjusted as necessary. The entire update process was completed in one afternoon hour last May.

I have added a module on professional applications in the field. This will involve a weeklong study during the middle of the term. My AI personal assistant reached out via emails it had found for the directors of HR at three regional firms who have hired many of our graduates in the past. The personal assistant briefed them on the course and got on their schedules to do in-class interviews via Zoom. All are on my schedule with times and connections. A reminder will be automatically sent out to the HR directors as well as put on my calendar and in the updated syllabus.  I will be asking the HR director about their hiring needs this coming year for new graduates as well as certificate completers from my university. Students will have a chance to pose some individual questions. I am hoping we will get into some serious career advice on where employment in my field is going. 

Students in the class will have access to their GenAI apps to formulate questions that are relevant to their own preferences and career choices. This is part of our department-wide emphasis on career development in an AI world. Also, part of that initiative is group work by the students using GenAI apps to lead their groups. Each group will report out with a collaboratively-written 10-page paper.

Fortunately, this fall I will have a version of the Khanmigo tutor that will answer students’ questions about assignments and discussion topics. Khanmigo will also help students with mini-lessons and quizzes to bring them up to speed in areas in which they fell short on prerequisites for the class. Khanmigo serves as a 24-hour smart tutor which has been trained on the course materials as well as other associated topics. The tutor, like the one developed by Khan Academy, is both empathetic and helpful.

Meanwhile my research project continues to percolate along with my personal AI agent sending out surveys, gathering data, crunching the numbers and applying advanced analytics to find previously-hidden correlations among demographic characteristics gathered from the samples. My AI agent is communicating with my co-author’s agent to ensure that we both are up to speed even though we live on different continents. The two agents are scanning journals and international conferences for the most appropriate venues for delivery of the final product of our research.

I am planning to go to the annual UPCEA conference in Denver in the spring. Today, I’ll ask my personal digital agent to gather all of the relevant registration, hotel and travel materials and give me a report to send to the dean’s office. I should get approval by the end of the week, and I’ll redirect that to the agent who will book the travel, hotel and registration. 

It’s so hard to keep up with the field, especially when you teach in communication technologies. Fortunately, my agent thoroughly scans all relevant articles and videos that come out from many hundreds of different sources, including research reports, journals, events coverage, YouTube videos, and associations. I get a daily, prioritized summary which I read each evening. I can extract entire articles with citations and put them in my very large cloud storage site. It’s easy to transfer AI-generated summaries to handouts for the class and put them into the ever-growing reading list for the class.

I am most thankful for the technology that spares me much of the pain of my five committee assignments. It seems a shame that faculty committees still run most universities. While we hate to surrender our oversight, the committees are the bane of so many faculty members, taking up far more time than seems worthwhile. Fortunately, our meeting software takes the minutes, advises on Roberts Rules of Order, summarizes the meetings, finds meeting times that work out best on all members’ calendars, and prepares the corrected minutes for the record.

As with so many of my colleagues, I am on the outlook for a new position where the financial conditions of the institution are stable, the community offers opportunities for my family, research support in my field is strong, and there is the opportunity to engage bright graduate students and postdocs into my field of research. Once again, it is my personal autonomous AI agent who is at work 24 hours a day seeking opportunities from the point of rumors on discussion boards to official position opening announcements. 

The above is what I anticipate will be reality for savvy faculty members in the fall of this year. Versions of the technology capabilities mentioned are already available. We are likely to see even more robust autonomous artificial agents than I describe. Artificial General Intelligence (AGI) is predicted to arrive this fall. And, some scientists predict Artificial Super-Intelligence will follow in 2027.

The technology is advancing at a much faster speed than most anyone predicted. It is propelled by unprecedented investments by many of the largest corporations around the world. The promise of massive profits for those who develop advanced marketable programs drives previously unexpected partnerships.

The message for those of us in higher education is that we must pay close attention to these developments for the sake of our students, our institutions and ourselves. 

This article was originally published in Inside Higher Ed’s Transforming Teaching & Learning blog.