Content uploaded by Bo Zhang
Author content
All content in this area was uploaded by Bo Zhang on Jan 24, 2023
Content may be subject to copyright.
1
Preparing Educators and Students for ChatGPT and AI Technology in Higher Education:
Benefits, Limitations, Strategies, and Implications of ChatGPT & AI Technologies.
Bo Zhang
University of Hartford
2
Abstract
Artificial intelligence (AI) has advanced significantly since it was developed and was
first integrated into higher education 60 years ago. Recently, the use of ChatGPT, a popular AI
tool released in November 2022, has sparked controversy among the higher education
community. The New York Department of Education announced a ban on access to ChatGPT
due to concerns that it may encourage students to violate academic integrity. This incident
caused a public conversation on the impact of technology and AI on educational institutions.
This paper aims to provide an overview of ChatGPT, including the background of ChatGPT, its
capabilities, benefits, and potential challenges and limitations. Additionally, this study addresses
the implications for educators and higher education institutions and offers recommendations for
addressing the concerns surrounding the use of ChatGPT and AI in higher education.
Introduction
Background
On the first week of January 2023, the New York Department of Education announced a
ban on access to a popular technology named ChatGPT, an artificial intelligence (AI) tool that
causes concerns about its potential encouragement of students to violate academic integrity
(Rosenblatt, 2023; Korn & Kelly, 2023). The news immediately raised public attention and an
unprecedented discussion on various social media and the internet about how technology and AI
have gradually and tremendously evolved and impacted education, specifically, teaching and
learning, and in the meantime, brought new issues and challenges that almost all global
3
educational institutions need to address at present (BBC News, 2023). However, in nearly the
same period, several other similar incidents were known are: New York Post reported that
assistant philosophy professor Darren Hick at Furman University caught one of his students
cheating by having ChatGPT write a 500-word essay as a take-home test at the end of the fall
semester 2022 (Mitchell, 2022); The student newspaper of Yeshiva University, The
Commentator, revealed that several students of the institution were reported for their academic
dishonesty related misconduct by cheating during a take-home exam during fall 2022 finals
using ChatGPT (Levin, 2023). BBC News (2023) reported that Stanford University already took
the initiative on recruiting lawyers to study the ChatGPT phenomenon seeking ways to address
and deal with the academic integrity issue.
The above cases occurred nearly the same time in late 2022 and early 2023, which was
merely several weeks after the ChatGPT was released to the public in November of the year. The
AI chatbot quickly enjoyed its popularity on social media and the internet (Levin, 2023;
Rosenblatt, 2023). However, what exactly is ChatGPT? How did it explode to the public and
caused big controversy?
ChatGPT and AI Technologies
ChatGPT is a “large language model” (LLM), an artificial intelligence system or tool that
is equipped with the ability to read, make summaries, solve math problems, produce course
syllabi drafts, tutor students and explain terms and abstract concepts and theories, find and fix
errors or bugs in the source code of computer programs; write various types essays (including
but not limited to: descriptive, narrative, persuasive, argumentative, and expository), create
consulting reports and new arts, recognize scholarly debates in various fields of study via its
access to relevant literature, provide translation of texts, create bibliographical references lists,
4
and predict words that will come next in a sentence that can create sentences comparable to what
human beings write and speak (Azaria, 2022; Mintz, 2023; Mollick, 2022; University of
Michigan, 2022). Developed by OpenAI
1
, GPT-2 was first released to the public in 2019 and
went viral in the fall of 2020 after the article “A Robot Wrote this Article: Are you scared yet
human?” was published in The Guardian via GPT-3. In the winter of 2020, Microsoft integrated
its editing function in Office 365 suit with the predictive text in GPT-3. By the end of 2022,
ChatGPT was open to the public, and it functions like a conversational agent, or chatbot,
providing services in generating content while interacting and helping its users with questions in
a variety of areas including academic study, finance, journalism, publishing, business, coding,
marketing, advertising, and consulting (Illinois State University, n.d.; Mollick, 2022). Needless
to say, the benefits and potential of ChatGPT have not yet fully unfolded to us. Similar AI-
generated content services are DALL-E, Jasper, and YouChat, Caktus ai, these are technologies
that are developed based on OpenAI (Illinois State University, n.d.).
The application of AI in higher education can be traced back to the 1960s when early
computer-assisted instruction (CAI) systems were developed and then employed in universities
(Dhawan & Batra, 2021). It has been playing an important role since then as many institutions
looking to integrate technology into their curriculums and institutional operations, especially
during the global COVID-19 pandemic with synchronous and asynchronous courses worldwide.
The growth of AI technologies in higher education has been propelled by factors such as
advances in computer hardware and software, the abundance of various types of data, and raised
funding for research in technology (Baker, 2021). Therefore, many AI-based tools and systems
1
OpenAI is a U.S. artificial intelligence research and deployment corporation that owns its for-profit company
OpenAI LP and its parent corporation, the non-profit OpenAI. The mission of OpenAI is to create and direct AI in
ways that are beneficial to all human beings (source: https://openai.com/).
5
have been developed, for instance, intelligent tutoring systems and virtual assistants such as
those roles in university libraries and helpdesk support offices.
ChatGPT is not the first technology that has been made available to the public. However,
unlike other technologies, it gained sudden fame soon after its release to the public while
immediately drawing concerns from the public as being viewed mostly as radical and negative.
As a current college student and an emerging researcher, I see the possibilities of
ChatGPT and AI technology in helping me in becoming more efficient in understanding theories
and concepts and illuminating ideas in writing thus improving my efficiency and productivity;
however, as an educator in an educational institution, I am concerned about how to evaluate
student’s work in an advanced technology era and better prepare and guide them in using this
new tool to support their academic growth and success while avoiding committing intellectual
dishonesty.
Challenges and Concerns
Different from earlier when people worry about being replaced by AI technology, the
need for increased surveillance and monitoring of students, and potential discrimination against
certain groups of people (Barron, 2003). As many recent articles argued, challenges and concerns
from the public associated with the use of AI in education are related to the issue of academic
integrity, plagiarism, the detection of the authenticity and originality of student work, and how to
take advantage of the technology and outweigh its drawbacks and limitations. Additionally,
ethical issues such as publishing, republication, and hiring and promotion in academia also need
to be addressed (Eaton et al., 2021). However, educators today “will likely continue to be
reactive rather than proactive as GPT-3 technologies evolve rapidly”; their values toward
assessment and evaluation of student work will need to change; the most frequently known and
6
used technologies that educators use need to be advanced; academic writing seems to be the most
concerned area for academic honesty; more students will use similar AI technologies for
efficiency and may not even aware it as academic integrity violation action; efforts need to be
made by institutions as for the effective approach to guiding educators and explicit policies to
enforce integrity (Eaton et al., 2021, p.18).
Advantages and Benefits
Regardless of its drawbacks, the advantages, and benefits that AI technologies, are
apparent at all academic levels, for example, it can make personalized learning available to
individual students providing customized instruction and services that meet student’s needs and
capabilities (Baker, 2021); it can collect and analyze various data from students such as their
performance and engagement and offer quick solutions for teaching and learning improvement
(Dhawan & Batra, 2021). For international students, the academic and financial potentials that
AI can bring are even more profound in terms of the cross-geographic location services and
instruction and time difference. In a world that pursues efficiency and productivity, AI such as
ChatGPT takes merely a few minutes of typing or simply copying and pasting your questions,
whether it is a mathematics expression, a college essay, a presentation, a business proposal, a
poem, or a research abstract (Else, 2023), a model answer or response is one-click away on the
mouse and a few seconds wait time. Speaking of its functions, it can (1) create paragraphs
writing in many areas and academic disciplines with a certain level of sophistication; (2)
generate fast and precise writing saving time and resources such as labor; (3) enable its users to
make corrections on its mistakes and can provide additional details in answers upon user request,
and (4) explain to users in response when being asked to do so (Azaria, 2022; Mollick, 2022).
7
Limitations of ChatGPT & AI Technologies
While we see the advantages and drawbacks, there are limitations of ChatGPT and other
text-generators in its performance as revealed in various articles:
• It may provide unreliable information on topics that has few citations.
• It is unable to alter words and use idioms smartly.
• It may produce inaccurate or false factual references.
• It is incapable of citing sources correctly and accurately.
• Its ability to produce responses weakens after several paragraphs.
• There is an absence of ethical considerations.
• It is unable to evaluate the quality and trustworthiness of sources.
• Logical errors can be found occasionally and may contradict themselves.
• Adjectives and adverbs-related grammatical errors exist.
• There is a potential bias lying within the technology such as it favors particular
numbers over others in mathematics.
• It has an inability in calculating the value of complex and difficult mathematical
expressions.
• It performs at a level of grade C to B- in writing assignments and other academic
fields of study.
• It is unaware of its lack of knowledge.
(Azaria, 2022; Eaton et al., 2021; Mintz, 2023).
Strategies for Instruction and Policy
As I read and explore the messages sent by many educators and administrators regarding
the recent emergence of ChatGPT and the discussion about AI technology, I was also looking for
8
strategies that educators may apply in their instruction, assessment, and student support. A
review of the literature found the following might be helpful:
1. Academic Study:
1) Promoting academic integrity among students
Possible strategies that serve this purpose are: include information about the honor
code, expectations, and the use of AI in the course syllabus; assign take-home essays;
emphasize and encourage participation; use in-class discussions, debates, oral
presentations, video or audio (i.e. podcasts) projects, and group projects that can
engage students for skill demonstration; add discussion boards; require multi-draft
writing; consider primary research assignments; create in-class writing assignments
using college-wide platforms such as Canvas (https://learn.canvas.net/login/canvas)
or BlackBoard (https://www.blackboard.com/); have student generate reflective
writing prompts or content based on instructional materials; encourage critical
thinking via various activities and assignments and incorporate them in assignment or
assessment; (Illinois State University, n.d.; Mitrano, 2023). Eaton et al. (2021) think
that some educators may also choose to return to traditional writing using paper and
pencil and may adopt AI technologies in using “automated grading” or “tighter
scaffolding” in class and assignments (p. 23). Educating students on ethical behaviors
in using technologies is also necessary for this matter (King, 2023).
2) Identifying AI-generated information
These strategies can be helpful: apply AI detection software such as Crossref
(https://www.crossref.org/); take the limitations listed above as a reference to identify
issues; pay attention when seeing a discrepancy in the writing style or language used
9
in different sections of an assignment; long passages without proper citations could be
an indication of plagiarism (Illinois State University, n.d.; King, 2023; Mitrano,
2023).
3) Engaging students in AI technologies
Provide opportunities in class to have students generate text using AI technologies
and then discuss and critique; allow students to create drafts via AI software and
practice revision and improvement; incorporate active learning activities such as
“think-pair-share” to expand learning experiences (Illinois State University, n.d.).
2. Student Support:
Mintz (2023) suggested using technologies such as ChatGPT for advising, career
counseling and writing instruction, and related feedback to students as those were areas
where resources or staff professional development opportunities are found oftentimes
deficient. Willsea (2023) indicated that it will be a trend that AI will be used to assist
college recruitment, marketing, and student engagement while decreasing costs and
overcoming staffing shortages.
Implications
Education in the past is to obtain knowledge, however, education now and in the future
may be how to search for information and or edit knowledge.
Technology and the internet have revolutionized legal research and writing. However,
there are limited research, practices, and policies have been found in addressing these issues,
which leads to a problem that needs higher education institutions’ immediate attention and
solution. With the abrupt emergence of ChatGPT and other AI language models, policymakers
10
should be proactive and forward-thinking in order to effectively address new challenges,
developments, and larger issues of ethics in education and technology (BBC News, 2023; Owoc
et al., 2021).
In sum, as Warner (2023) and Mintz (2023) pointed out, similar to the introduction of
calculators in math classes and mechanical looms and knitting frames to the British textile
industry in the 19th century, surveilling, punishing, or banning ChatGPT is not a good solution as
we think about modern education. While the output may be similar, the underlying labor is
different, not to mention writing itself is a form of thinking and exploration. Higher education
institutions should not perceive it as simply a crisis or challenge and take the censorship route
but should instead see the possibilities and focus on how to maintain academic integrity while
allowing students to use AI wisely as a resource for innovation to excel in machine learning
beyond its limitations for success (Mitrano, 2023).
In other words, a better solution would be to put learning at the center and give students
something they believe is worth doing. It is time for us to think about how to continue the
discussion of using ChatGPT and AI technology with an emphasis on underlying values in
education just like how we accept and benefit ourselves from Google with its memory and
complex calculations and Wikipedia as an electronic library.
11
References
Azaria, A. (2022). ChatGPT usage and limitations. ResearchGate.
DOI:10.13140/RG.2.2.26616.11526
Baker, R. S. (2021). Artificial intelligence in education: Bringing it all together. In OECD
Digital Education Outlook 2021: Pushing the Frontiers with Artificial Intelligence,
Blockchain and Robots, (pp. 43-54). OECD
Publishing. https://doi.org/10.1787/f54ea644-en.
BBC News. (2023, January 15). What is ChatGPT, the AI software taking the internet by storm?
[Video] BBC News. https://www.youtube.com/watch?v=BWCCPy7Rg-s
Dhawan, S. & Batra, G. (2021). Artificial intelligence in higher education: Promises, perils, and
perspectives. ResearchGate.
https://www.researchgate.net/publication/348910302_Artificial_Intelligence_in_Higher_
Education_Promises_Perils_and_Perspective
Eaton, A. E., Mindzak, M. & Morrison, R. (2021, June). The impact of text-generating
technologies on academic integrity: AI & AI. ResearchGate.
https://www.researchgate.net/publication/353169564_The_impact_of_text-
generating_technologies_on_academic_integrity_AI_AI
Else, H. (2023, January 12). Abstract written by ChaptGPT fool scientists. Nature.
https://www.nature.com/articles/d41586-023-00056-7
Illinois State University. (n.d.). AI-generated content in the classroom: Considerations for
course design. https://prodev.illinoisstate.edu/pedagogy/ai/
12
King, M. R. (2023, January 2). A conversation on artificial intelligence, chatbots, and plagiarism
in higher education. Cel. Mol. Bioeng. 16, 1–2 (2023). https://doi.org/10.1007/s12195-
022-00754-8
Korn, J. & Kelly, S. (2023, January 6). New York City public schools ban access to AI tool that
could help students cheat. CNN. https://www.cnn.com/2023/01/05/tech/chatgpt-nyc-
school-ban/index.html
Levin, J. (2023). Student caught cheating using AI on final; Academic integrity policy updated.
The Commentator. https://yucommentator.org/2023/01/students-caught-cheating-using-
ai-on-final-academic-integrity-policy-updated/
Mintz, S. (2023, January 16). ChatGPT: Threat or Menace? Inside Higher Ed.
https://www.insidehighered.com/blogs/higher-ed-gamma/chatgpt-threat-or-menace
Mitchell, A. (2022, December 26). Professor catches student cheating with ChatGPT: ‘I feel
abject terror’. New York Post. https://nypost.com/2022/12/26/students-using-chatgpt-to-
cheat-professor-warns/
Mitrano, T. (2023, January 17). Coping with ChatGPT. Inside Higher Ed.
https://www.insidehighered.com/blogs/law-policy%E2%80%94and-it/coping-chatgpt
Owoc, M.L., Sawicka, A., Weichbroth, P. (2021). Artificial Intelligence Technologies in
Education: Benefits, Challenges and Strategies of Implementation. In: Owoc, M.L.,
Pondel, M. (Eds.), Artificial Intelligence for Knowledge Management. AI4KM 2019. IFIP
Advances in Information and Communication Technology (pp.37-58). Springer. DOI:
10.1007/978-3-030-85001-2_4
13
Rosenblatt, K. (2023, January 5). ChatGPT banned from New York City public schools’ devices
and networks. NBC News. https://www.nbcnews.com/tech/tech-news/new-york-city-
public-schools-ban-chatgpt-devices-networks-rcna64446
University of Michigan. (2022, November 7). Parthasarathy discusses implications of Large
Language Models. https://fordschool.umich.edu/news/2022/parthasarathy-discusses-
implications-large-language-models
Waner, J. (2023, January 16). ChatGPT both is and is not like a calculator. Inside Higher Ed.
https://www.insidehighered.com/blogs/just-visiting/chatgpt-both-and-not-calculator
Willsea, M. (2023, January 20). Top 3 challenges facing higher education Marcomm in 2023.
Inside Higher Ed. https://www.insidehighered.com/blogs/call-action-marketing-and-
communications-higher-education/top-3-challenges-facing-higher