To provide a basic overview of generative artificial intelligence and how it can be used in higher education settings


Generative artificial intelligence (AI) includes technologies that provide responses based on the user typing in a specific prompt. Responses are typically text output, an image, artwork, or other media. The output is generated using algorithms that synthesize information from large datasets of information. Generative AI technologies are also called large language models (LLMs) and give the feel that the computer is in essence, talking back. There are a variety of usages of these tools in higher education, and over time they continue to become more sophisticated. 

Some sample AI tools include ChatGPT-3/ChatGPT-4, chatPDF, Dall-E-2, Elicit, and Google Bard. Please note this is a list and does not endorse any particular tool. Some tools are openly accessible and others have a fee. As the technologies evolve, more generative AI tools continue to be released, and they are increasingly embedded within existing applications and are being used in higher education. In a social media analysis of early adopters, Mogavi et al. (2023) found that ChatGPT was used for work such as content creation and editing (most prevalent), content summarization, collaborative data analysis, and student evaluation. 

A natural response to the expansion and sophistication of these technologies is concern over academic integrity. These tools will likely continue to envelop critical spaces within our educational systems and it is imperative that instructors are equipped to use them and that students understand how they can be leveraged to support their learning. Additionally, staff within higher education may find these tools can help support their work by building efficiencies. We are at a time period when it is important for all of higher education to develop AI literacy. To this end, this resource provides some guidance on the usage of AI in teaching and learning, research, and administrative tasks.

Teaching and Learning

Instructor Work

Large language models can save valuable instructional preparation time and grading. For example, instructors can use the tools to create draft syllabi for a new course, prepare lecture notes, as well as create slides for a presentation. Additionally, such tools can be used to create a grading rubric for a particular assignment, or an assessment for specific course content. 

Student Work 

In general, all courses should have explicit policies around the usages of generative AI that are communicated to students for each assignment (examples forthcoming). Ideally instructors should hold a classwide discussion about the policy to allow for more in depth conversations.

Students can use LLMs as study assistants or, depending on course policies, for assignment completion. For example, for study purposes, students can upload open access course readings into the technologies and generate summaries to help them break down the topics or concepts better. Learners can also use them to create practice assessments to test themselves on the course material and practice research-supported study strategies such as retrieval or concept mapping. 

For assignment creation, students can use the tools to brainstorm topics for a paper or project, create thesis statements, develop outlines for writing assignments, as well as for proofreading. Students can use the tools to develop or improve simple code for programming courses. Generative AI technologies can be quite advantageous for students to proofread their work, for neurodivergent students with ADHD to actively process course content, as well as students with specific learning disabilities as assistive technologies. 

When students use AI in assignment creation instructors should be sure to request copies of their outputs at each stage and an indication of which tools they used in completing the assignment. Such documentation allows for a better understanding of how the students completed the work, and potentially provides opportunities for students to reflect on how the tools supported their process.


The usage of AI in scholarly endeavors has much potential. For tools that allow the upload of a .pdf file, summaries of open access academic manuscripts can be generated to synthesize the content. Such overviews may have utility when conducting a literature review or participating in a journal club or discussion.  

Additionally, the tools can be used to help draft and proofread manuscripts, or create an annotated bibliography or literature review. The outputs should always be viewed with caution when used for these purposes as they might not produce the correct sources. All references should be double-checked for accuracy. To find books and articles, Google Scholar and library databases like OneSearch or Web of Science are recommended over AI tools.

Some are finding potential usage of AI tools in performing data analyses, effectively using them as research assistants. Scholars have imputed textual data into the tools and asked them to perform annotation, develop themes, and conduct other forms of data analysis (Mogavi et al., 2023). Another application of using these tools is generating ideas for more compelling grant proposals. 

There are ethical implications in using AI in scholarly endeavors. For example, articles, books, and grant proposals submitted through peer-review processes are typically to be held confidential and the use of generative AI technologies for their evaluation or review can be a breach of such confidentiality. Ensure through the proper channels that imputing such data is acceptable. 

Administrative Work 

As AI continues to become increasingly embedded into existing software tools such as those for email, word processing, and presentation applications, usage may continue to expand. 

Some tools integrate with email clients and can help deliver a clearer message by providing suggested words or phrases. The technologies can also be used to develop an initial draft of an email message that is tailored to the receiver. Similar benefits are possible with word processing tools. 

Other administrative work such as annual reports, memos, reference letters, can also be proofread within these tools to produce higher quality writing. Additionally, sample templates or outlines of each of the previously mentioned types of documents can be created through generative AI and used as part of the prewriting process. 

Generative AI can be used in graphics design, enabling the creation of images through text responses, and bypassing the need to learn how to use sophisticated design software. Additionally, AI is increasingly integrated within applications that generate sample alt text to convey information to individuals with visual impairments.  

Cautionary Areas

There are many possible benefits to using generative AI in higher education, but also some cautions: 

  • Fact-checking: Because LLMs pull from large datasets which might be of varying quality, some responses might be inaccurate. A critical eye should be applied to all output.
  • Bias: Outputs might contain biased language, perspectives, or display linguistic bias. A number of sites indicate such disclaimers. 
  • Personalization: The outputs of generative AI can be useful starting points for a variety of work, but they might not be appropriately tailored towards the particular objectives and require substantial revision. 
  • Feeding the tool: The more data imputed into the tools, the more the applications are fueled with more content. In using these technologies, one should consider the pros and cons of this scenario and what they choose to impute.
  • Access: At any point open access or free tools could switch to paid subscriptions, the pricing of which can increase over time.
  • Privacy: Any data imputed could become part of the public dataset from which the tool pulls, thus it is important not to submit private or confidential information.
  • Ethics: Some usages of generative AI can have ethical implications and might be considered breaches of confidentiality. 

The usage of generative AI technologies in some teaching and learning; research; and administrative activities holds promise in higher education. As these technologies continue to be integrated into everyday experiences, students, faculty, and staff members will likely continue to consider their utility and weigh their cautionary areas. 


Mogavi et al. (preprint, 2023). Exploring User Perspectives on ChatGPT: Applications, Perceptions, and Implications for AI-Integrated Education.