AI Policy Education ChatGPT Policy in HE

Chan Int J Educ Technol High Educ (2023) 20:38 International Journal of Educational
https://doi.org/10.1186/s41239-023-00408-3
Technology in Higher Education
RESEARCH ARTICLE Open Access
A comprehensive AI policy education

framework for university teaching and learning
Cecilia Ka Yuk Chan1*
*Correspondence:
[email protected] Abstract
1
University of Hong Kong, Hong This study aims to develop an AI education policy for higher education by examining
Kong, Hong Kong, HKSAR, China the perceptions and implications of text generative AI technologies. Data was collected
from 457 students and 180 teachers and staff across various disciplines in Hong Kong
universities, using both quantitative and qualitative research methods. Based on the
findings, the study proposes an AI Ecological Education Policy Framework to address
the multifaceted implications of AI integration in university teaching and learning. This
framework is organized into three dimensions: Pedagogical, Governance, and Opera-
tional. The Pedagogical dimension concentrates on using AI to improve teaching and
learning outcomes, while the Governance dimension tackles issues related to privacy,
security, and accountability. The Operational dimension addresses matters concern-
ing infrastructure and training. The framework fosters a nuanced understanding of
the implications of AI integration in academic settings, ensuring that stakeholders are
aware of their responsibilities and can take appropriate actions accordingly.
Highlights
• ProposedAI Ecological Education Policy Framework for university teaching and

learning.
• Threedimensions: Pedagogical, Governance, and Operational AI Policy Frame-
work.
• Qualitativeand quantitative data collected from students, teachers, and staff.
• Tenkey areas identified for planning an AI policy in universities.
• Studentsshould play an active role in drafting and implementing the policy.
Keywords: AI policy framework, Artificial intelligence, ChatGPT, Ethics, Assessment
Introduction
In recent months, there has been a growing concern in the academic settings about the
use of text generative artificial intelligence (AI), such as ChatGPT, Bing and the latest,
Co-Pilot integrated within the Microsoft Office suite. One of the main concerns is that
students may use generative AI tools to cheat or plagiarise their written assignments
and exams. In fact, a recent survey of university students found that nearly one in three
© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits
use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original
author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third
party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the mate-
rial. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or
exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://
creativecommons.org/licenses/by/4.0/.
Chan Int J Educ Technol High Educ (2023) 20:38 Page 2 of 25
students had used a form of AI, such as essay-generating software, to complete their
coursework (Intelligent.com, 2023). About one-third of college students surveyed (sam-
ple size 1000) in the US have utilized the AI chatbot such as ChatGPT to complete writ-
ten homework assignments, with 60% using the programme on more than half of their
assignments. ChatGPT types of generative AI tools is capable of imitating human writ-
ing, with some students using it to cheat. The study found that 75% of students believe
that using the programme for cheating is wrong but still do it, and nearly 30% believe
their professors are unaware of their use of the tool. The study also noted that some pro-
fessors are considering whether to include ChatGPT in their lessons or join calls to ban
it, with 46% of students saying their professors or institutions have banned the tool for
homework. This has led to calls for stricter regulations and penalties for academic mis-
conduct involving AI.
Another concern is that the use of generative AI may lead to a decline in students’
writing and critical thinking skills (Civil, 2023; Warschauer et al., 2023), as they become
more reliant on automated tools to complete their work. Some academics argue that
this could have a negative impact on the quality of education and ultimately harm the
students’ learning outcomes (Chan & Lee, 2023; Korn & Kelly, 2023; Oliver, 2023; Zhai,
2022).
These concerns have led some universities to ban the use of generative AI in their aca-
demic programmes. Eight out of 24 universities in the prestigious UK Russell Group
have declared the use of the AI bot for assignments as academic misconduct including
Oxford and Cambridge. Meanwhile, many other universities around the world, are rush-
ing to review their plagiarism policies citing concerns about academic integrity (Wood,
2023; Yau & Chan, 2023). Some Australian universities have had to alter their exam and
assessment procedures back to pen- and paper-based (Cassidy, 2023; Cavendish, 2023).
However, there are also those who argue that generative AI has the potential to revolu-
tionize education and enhance the learning experience for students. For example, some
experts suggest that generative AI could be used to provide personalized feedback and
support to students, helping them to identify areas of weakness and improve their skills
in an adaptive manner (Kasneci et al, 2023; Sinhaliz et al., 2023).
Generative AI and generative pre‑trained transformers

Generative AI is a subset of artificial intelligence (AI) that focuses on creating new data
or content rather than analysing and interpreting existing data (McKinsey Consultant,
2023). Generative Pre-trained Transformers (GPT) are a type of generative AI model
that use deep learning techniques to generate natural language text. The latest versions
of GPT, GPT-3.5 and GPT-4, large language models which are trained on a large cor-
pus of text data, are capable of producing human-like text with high levels of coherence,
complexity, and diversity. GPT-3.5 and GPT-4 are both examples of artificial general
intelligence (AGI), which is the ability of AI systems to perform any intellectual task that
a human can do. Unlike artificial narrow intelligence (ANI), which is designed to per-
form a specific task, AGI is designed to perform multiple tasks and generalize knowl-
edge across different domains. While GPT-3.5 and GPT-4 are not true AGI systems, they
represent significant progress towards achieving AGI by demonstrating the ability to
perform a wide range of language tasks and generate human-like text. The development
of generative AI models like GPT-3.5 and GPT-4 has the potential to revolutionize many
fields, including natural language processing, creative writing, and content generation.
Rationale for an artificial intelligence education policy

With generative AI tools becoming easily accessible to the public in recent months, they
are rapidly being integrated into various fields and industries. This has created an urgent
need for universities to develop an AI education policy that prepares students to work
with and understand the principles of this technology. There are several rationales sup-
porting this need.
Firstly, AI technology is becoming more prevalent in many sectors of the economy,
such as finance (Bholat & Susskind, 2021; Buckley et al., 2021), healthcare (Eggmann
et al., 2023; Yu et al., 2018), and transportation (Abduljabbar et al., 2019; Wu et al., 2022).
As a result, graduates will need to have a strong understanding of AI principles in order
to succeed in these fields. An AI education policy can provide students with the neces-
sary knowledge and skills to work with AI in a professional capacity.
Secondly, AI has the potential to revolutionize many aspects of society, including edu-
cation itself (Adiguzel et al., 2023). AI can be used to enhance student learning by pro-
viding personalized, real-time feedback and adapting to individual learning styles (Atlas,
2023; Chan & Hu, 2023; Luckin, 2017). By educating students on AI, universities can
help prepare them to be active participants in the development and implementation of
AI technology, ensuring that it benefits society as a whole.
Thirdly, as the use of AI in education and assessment becomes more prevalent, it is
essential that students understand the principles behind the technology in order to
maintain academic integrity and prevent cheating as mentioned previously (Chan,
2023; Cotton et al., 2023). An AI education policy can teach students about the ethical
considerations surrounding AI, such as bias and fairness, as well as the potential conse-
quences of using AI in academic contexts.
Fourthly, developing an AI education for university is important to prepare students
for the future (Aoun, 2017). AI technology is rapidly advancing, and it is likely to play
an increasingly important role in society in the coming years. By providing students and
teachers with training in AI, universities can help ensure that graduates are equipped to
contribute to the development of AI and to navigate the ethical, social, and economic
issues that are likely to arise as AI becomes more widespread. Such training should also
help students become competent and responsible users of AI in their daily lives.
Finally, it is also worth mentioning that previous AI policies in education did not antic-
ipate the level of advancements that text-based GPT 3.5 and 4 can now achieved. Given
the potential benefits and risks associated with the use of generative AI in education, it
is important to develop a proper AI education policy that addresses these concerns and
provides guidance on the responsible use of AI.
Conducting research in AI policy in education within Hong Kong is specifically jus-
tified due to the city’s unique position as a global hub of technology, commerce, and
education, coupled with its evolving education landscape. As a dynamic metropolis
with a strong commitment to technological innovation and a richly diverse education
system, Hong Kong presents a compelling case study for the exploration of AI policies
in education. Its blend of Eastern and Western educational philosophies and practices
offers a fertile ground for examining the impacts and opportunities of AI integration in
varied educational contexts. Furthermore, as Hong Kong is actively striving to enhance
its digital learning capabilities and infrastructure, studying AI policy could provide val-
uable insights into the challenges and best practices of implementing AI in education,
thereby potentially informing AI education strategies not only in Hong Kong but also in
other global contexts.
The study employed a comprehensive approach to data collection, gathering rich
quantitative and open-ended survey data from a diverse range of stakeholders in the
education community to ensure that it reflects the needs and values of all those involved.
The combination of these data sources allowed for a holistic understanding of the topic
under investigation, providing a nuanced and multifaceted view of the issues at hand. By
doing so, we can help to ensure that the use of generative AI in education is both benefi-
cial and ethical.
Existing policy on artificial intelligence

The aim of this study is to investigate the education policy related to AI, however, it is
essential to also scrutinize the existing policies governing AI as a whole. As AI expands
its sphere of influence to various sectors in our society, there are increasing concerns
over the risks of its usage and how it might impact human activities (AI regulation,
2023; World Economic Forum [WEF], 2023). Some of the major issues of concern that
have drawn the attention of governments around the world include discrimination and
bias of AI, loss of privacy, violation of human rights, and malicious use of AI (Greiman,
2021; Hogenhout, 2021). Federspiel et al. (2023) caution that misuse of AI could encour-
age manipulation of people, create social division, and exacerbate inequalities, posing
existential threats to the human race. In view of this, countries have been working on
national policies and strategies to provide clearer guidance on AI usage in order to maxi-
mize its benefits while mitigating the threats brought by it.
To advocate the responsible and proper management of AI technologies, the centre
of focus for most national policies on AI have fallen on the discussion of ethics, which
deals with “the standards of right and wrong, acceptable and not acceptable” (Hogen-
hout, 2021, p. 11). Floridi (2021)’s framework for the ethical use of AI, which proposed
the 5 core principles of “beneficence, non-maleficence, autonomy, justice and explica-
bility”, is referred to by most national policies on AI as a foundation to further develop
on. In addition, Dexe and Franke (2020) summarized the AI strategy documents from
the Nordic countries and identified various ethical principles as the implicit foundation
for further developing policies. The official AI governance framework from Singapore
also recognized the “explainable, transparent and fair usage of AI in decision-making
process” and “human-centric AI solutions” as the guiding principles of ethical use of AI
(IMDA & PDPC, 2020). Apart from individual countries, ethics has been the empha-
sis of the AI policies published by regional and international bodies. UNESCO devel-
oped its guidelines on the ethical use of AI technologies by emphasizing the key idea
of human-centeredness and hence, human rights and values laid out in the Universal
Declaration of Human Rights (UDHR) are advised to be adopted as the necessary foun-
dation to further promote beneficial and appropriate use of AI technologies (UNESCO,
2021b, 2023). AI strategy in the European Union, as Renda’s (2020) analysis pointed out,
also focused on ethics and highlighted a human-centric approach to AI. In order to pro-
tect EU citizens from the danger of abusive use of advanced technologies, EU proposed
its own pillars (legal compliance, ethical alignment and sociotechnical robustness) to
ensure the trustworthiness of AI and established a specific AI expert group to work on
specific policy recommendations and guidelines.
The heavy focus that these national and regional policies has placed on ethics dem-
onstrates how limited they can do for the implementation of AI technologies. On the
one hand, difficulty to lay down a universal definition on ethical principles becomes a
hinderance for certain countries in formulating policies on the use of AI (Dexe & Franke,
2020). On the other hand, as AI can weave into the fabrics of everyday human activi-
ties, the resulting wide coverage of policy areas ranging from governance to education
and even to environment makes it a challenging task for government to establish specific
policies on AI usage (UNESCO, 2021b). Thus, as the Singaporean AI governance frame-
work highlighted, model framework or ethical guidelines were in themselves directional
and for reference only, and AI practitioners need to consider them with flexibility and
according to the relevance of particular situations (IMDA & PDPC, 2020).
Moving forward, the ongoing efforts of national and international organizations to
ensure the positive implementation of AI technologies will continue to prioritize discus-
sions and the formulation of legal and ethical principles (AI regulation, 2023; UNESCO,
2023). However, until these principles are validated by real-time implementation of AI
technologies, they will remain primarily predictive and prescriptive in nature (Chatter-
jee, 2020). Over time, it may become necessary for countries to establish institutional
support systems to effectively manage AI practices in accordance with validated legal
and ethical guidelines (Renda, 2020).
Table 1 Compilation of fundamental ethical principles (IMDA &PDPC, 2020)

Fundamental ethical principles for AI
1. Accountability: Ensure AI actors are held responsible for the AI systems’ functioning and adherence to ethical
principles
2. Accuracy: Recognize and communicate sources of error and uncertainty in algorithms and data to inform
mitigation procedures
3. Auditability: Allow third parties to examine and review algorithm behavior through transparent information
disclosure
4. Explainability: Ensure that algorithmic decisions and underlying data can be explained in layman’s terms
5. Fairness: Prevent discriminatory impacts, include monitoring mechanisms, and consult diverse perspectives
during system development
6. Human Centricity and Well-being: Prioritize the well-being and needs of humans in AI development and
implementation
7. Human rights alignment: Ensure technologies do not violate internationally recognized human rights
8. Inclusivity: Make AI accessible to everyone
9. Progressiveness: Favour projects with significantly greater value than their alternatives
10. Responsibility, accountability, and transparency: Build trust through responsibility, accountability, and fairness,
provide avenues for redress, and maintain records of design processes
11. Robustness and Security: Ensure AI systems are safe, secure, and resistant to tampering or data compromise
12. Sustainability: Favour implementations that provide long-lasting, beneficial insights and can predict future
behavior
Below consists of a compilation of fundamental ethical principles for AI that have been
extracted from multiple policies (IMDA & PDPC, 2020) (Table 1).
Existing policy on AI in education

The integration of AI technologies into teaching and learning has begun as early as
the 1970s and nowadays, different forms of these technologies are used in various
educational contexts, such as the use of personalized applications for learning and
assessment and information systems that help handle administrative and manage-
ment tasks in schools (Al Braiki et al., 2020; Schiff, 2022; UNESCO, 2021a).
As mentioned above, the use of AI technologies has raised different issues of con-
cern (Chan & Tsi, 2023; Chan & Zhou, 2023). In the educational contexts, other
than the general risks brought by the use of AI, concerns are primarily centred on
issues such as what changes can AI bring to the design of assessment and curriculum,
equalities and universality in accessing these technologies, redefinition of teachers’
role, and the lack of technological infrastructure for emerging economies (Pelletier
et al., 2022; Popenici & Kerr, 2017; Swiecki et al., 2022; TEQSA, 2023; UNESCO,
2021a). Based on these concerns, AI policies in education fix their eyes on address-
ing a number of issues: literacy education to prevent inequalities in the use of digi-
tal technologies (Southgate, 2020; UNESCO, 2021b); essential values of traditional
forms of teaching and learning such as teacher-student and student–student relation-
ships (Luan et al., 2020; UNESCO, 2021b); inclusiveness and equity in the use of AI
technologies (Tanveer et al., 2020; UNESCO, 2021a); professional development of
teachers to enhance teachers’ competence and assist them to transform their roles
(Ocaña-Fernández et al., 2019; Wang et al., 2021); and training and enhancement of
skills or “micro-credentials” for students that are important and necessary for har-
nessing technologies (Pelletier et al., 2022; UNESCO, 2021a). The roles of literacy
education and skills training are having particular implications for the wider society
as the population in general also needs to be prepared for the implementation of AI
technologies in different sectors.
Despite identifying multiple issues of concern in the educational contexts, policies
on AI in education are mostly generic and implicit because of the lack of concrete evi-
dence of implementing AI technologies (UNESCO, 2021a). In Schiff (2022)’s review on
24 AI policy strategies focusing on the role of education in global AI policy discourse, it
was found that policymakers view education largely as an instrumental tool to support
workforce development and training of AI experts. The article finds that the use of AI
in education is largely absent from policy conversations, while the instrumental value
of education in supporting an AI-ready workforce and training more AI experts is over-
whelmingly prioritized. The article suggests that if such a trend continues, policymakers
may fail to realize AI in education’s transformative potential and may fail to sufficiently
fund, regulate, and consider AI in education’s ethical implications. AI scholarship and
education governance do not receive adequate attention in the current literature (Gellai,
2022), and public understanding of the policy implications of AI is limited (Feldstein,
2019). While more work is still to be done in order to formulate more comprehensive
and focused policy documents on AI in education, ethics was reiterated again as a stra-
tegically plausible starting point for further discourse and researchers were especially
encouraged to engage further with policymakers through their work on ethics in the use
of AI in education (Sam & Olbrich, 2023; Schiff, 2022). In view of this gap, this research
intends to propose a policy framework for integrating AI in higher education, taking into
consideration aspects of teaching and learning as well as ethical and practical concerns.
In this research, we will employ the guidelines put forth by UNESCO (2021a) as
the starting point for crafting a more accurate AI policy for university teaching and
learning. The rationale for employing UNESCO recommendations as the basis is mul-
tifaceted. First, UNESCO is an esteemed international organisation with significant
expertise in education, their recommendations are supported by thorough research
and knowledge from experts worldwide. These recommendations are designed to be
relevant and flexible for a variety of educational systems and cultural settings, making
them suitable for diverse institutions. UNESCO’s guidelines also take a comprehen-
sive approach to incorporating AI in education, addressing important ethical, social,
economic, and technological aspects essential for creating effective policies. Using an
existing framework like UNESCO’s recommendations saves time and resources and
provides a well-organized starting point for examining specific AI policy issues in
university teaching and learning. Finally, anchoring the study in UNESCO’s recom-
mendations enhances the credibility of the research.
The UNESCO framework for AI in education is centred around a humanistic
approach, which aims to safeguard human rights and provide individuals with the
necessary skills and values for sustainable development, as well as effective human–
machine collaboration in life, learning, and work. The framework prioritizes human
control over AI and ensures that it is utilized to improve the capabilities of both
teachers and students. Moreover, the framework calls for ethical, transparent, non-
discriminatory, and auditable design of AI applications. From the UNESCO’s AI and
Education: Guidance for Policy-Makers document, the following recommendations
are provided:
1. Interdisciplinary planning and inter-sectoral governance: This recommendation

suggests that AI and education policies should be developed through collaboration
between different sectors and disciplines to ensure a comprehensive approach. For
example, policymakers could work with experts in education, technology, ethics, and
other relevant fields to develop policies that take into account all aspects of AI in
education.
2. Policies on equitable, inclusive, and ethical use of AI: This recommendation empha-
sizes the importance of ensuring that AI is used in an ethical and inclusive man-
ner that benefits all learners. For example, policymakers could develop policies that
address issues such as bias in AI algorithms or access to AI tools for learners from
disadvantaged backgrounds.
3. Develop a master plan for using AI for education management, teaching, learning,
and assessment: This recommendation suggests that policymakers should develop a
comprehensive plan for using AI in various aspects of education to ensure its effec-
tive implementation. For example, a master plan could include specific goals for
using AI in areas such as personalized learning or teacher professional development.
4. Pilot testing, monitoring and evaluation, and building an evidence base: This rec-
ommendation highlights the importance of testing and evaluating the use of AI in
education through pilot projects to build an evidence base for its effectiveness. For
example, policymakers could fund pilot projects that test the use of AI tools in spe-
cific educational contexts or with specific learner populations.
5. Fostering local AI innovations for education: This recommendation suggests that
policymakers should encourage the development of local innovations in AI for edu-
cation to ensure that it meets the specific needs of their communities. For example,
policymakers could provide funding or support to local startups or research institu-
tions working on developing new AI tools or applications specifically designed for
their region’s educational needs.
Using UNESCO’s recommendations as a basis, this study aims to examine higher edu-
cation stakeholders’ perceptions of text generative AI technology. Based on their ideas,
recommendations, and concerns, an AI education policy framework will be developed
to promote ethical and effective integration of AI technologies in higher education.
Methodology
In this study, a survey design was utilized to gather data from students, teachers, and
staff in Hong Kong to develop AI education policy framework for university teaching
and learning. The survey was administered through an online questionnaire, featuring a
mix of closed-ended and open-ended questions. The questionnaire was designed based
on a review of current literature on AI use in higher education. Topics covered in the
survey were major issues concerning the use of AI in higher education, which included
the use of generative AI technologies like ChatGPT, the integration of AI technologies
in higher education, potential risks associated with AI technologies, and AI’s impact on
teaching and learning.
Data were collected via an online survey from a diverse group of stakeholders in the
education community, ensuring that the results reflect the needs and values of all par-
ticipants. A convenience sampling method was employed for selecting the respondents,
based on their availability and willingness to participate in the study. Participants were
recruited through an online platform and provided with an informed consent form prior
to completing the survey.
The survey was completed by 457 undergraduate and postgraduate students, as well
as 180 teachers and staff members across various disciplines in Hong Kong. Descrip-
tive analysis was used to analyse the survey data, while a thematic analysis approach was
applied to examine the responses from the open-ended questions in the survey.
Quantitative data (survey data) and descriptive analysis

A range of survey items was included to capture different aspects of participants’ usage
and perception of generative AI technologies like ChatGPT. For example, participants
were asked whether they have used ChatGPT or similar generative AI technologies
before and how they envision using these technologies in their teaching and learning
practices.
Descriptive analysis was employed to analyse the survey data collected from students
and teachers in Hong Kong, in order to gain a better understanding of the usage and
perception of generative AI technologies like ChatGPT in higher education. Descriptive
analysis is an appropriate statistical method for summarizing and describing the main
characteristics of the sample and the data collected. It is particularly useful for analysing
survey data and can provide an overview of the distribution, central tendency, and vari-
ability of the responses.
Qualitative data (open‑ended data) and thematic analysis

Aside from the quantitative part of the survey, respondents were also asked about their
apprehensions regarding the integration of generative AI technologies and their rec-
ommendations for university strategic plans through open-ended questions. Different
perspectives and experiences were gathered from the respondents across various disci-
plines. The data from the open-ended questions were analysed using a thematic analy-
sis approach, which involved identifying patterns and themes in the data. An inductive
approach was used to analyse the responses, where the themes emerged from the data
rather than being predetermined by the researcher.
The combination of the quantitative and qualitative data enabled a more holistic
understanding of the usage and perception of generative AI technologies in higher edu-
cation. This allowed for the pinpointing of potential requirements, recommendations,
and strategies for AI policy in university teaching and learning. This understanding is
essential for ensuring that the use of these technologies is both beneficial and ethical.
Results
Findings from the quantitative data
The survey was conducted among 457 students and 180 teachers and staff from different
disciplines in Hong Kong universities. The goal was to explore the kinds of requirements,
guidelines and strategies necessary for developing AI policies geared towards university
teaching and learning. The findings reveal valuable insights into the perception of gen-
erative AI technologies like ChatGPT among students and teachers (refer to Table 2).
Regarding the usage of generative AI technologies, both students (mean = 2.28,
SD = 1.18) and teachers (mean = 2.02, SD = 1.1) reported relatively low experience, sug-
gesting that there is significant room for growth in adoption. Both groups demonstrated
a belief in the positive impact of integrating AI technologies into higher education (stu-
dents: mean = 4, SD = 0.891; teachers: mean = 3.87, SD = 1.32). This optimism was also
reflected in the strong agreement that institutions should have plans in place associ-
ated with AI technologies (students: mean = 4.5, SD = 0.854; teachers: mean = 4.54,
SD = 0.874).
Both students and teachers were open to integrating AI technologies into their
future teaching and learning practices (students: mean = 3.93, SD = 1.09; teachers:
mean = 3.92, SD = 1.31). However, there were concerns among both groups about other
students using AI technologies to get ahead in their assignments (students: mean = 3.67,
SD = 1.22; teachers: mean = 3.93, SD = 1.12). Interestingly, both students and teachers
did not strongly agree that AI technologies would replace teachers in the future (stu-
dents: mean = 2.14, SD = 1.12; teachers: mean = 2.26, SD = 1.34).
Table 2 Descriptive analysis for quantitative results

Item Students Teachers
N Mean Median SD N Mean Median SD
I have used generative AI technologies like 457 2.28 2 1.18 180 2.02 2 1.1
ChatGPT
The integration of generative AI technologies like 457 4 4 0.891 180 3.87 4 1.32
ChatGPT in higher education will have a positive
impact on teaching and learning in the long run
Higher education institutions should have a plan 457 4.5 5 0.854 180 4.54 5 0.874
in place for managing the potential risks associ-
ated with using generative AI technologies like
ChatGPT in teaching and learning
I envision integrating generative AI technolo- 455 3.93 4 1.09 180 3.92 4 1.31
gies like ChatGPT into my teaching and learning
practices in the future
I am concerned that other students may use gen- 456 3.67 4 1.22 180 3.93 4 1.12
erative AI technologies like ChatGPT to get ahead
in their assignments. /I am concerned that there
may be an unfair advantage for some students
as they may use generative AI technologies like
ChatGPT to get ahead in their assignments
AI technologies like ChatGPT will replace teach- 457 2.14 2 1.12 180 2.26 2 1.34
ers in the future
Students must learn how to use generative AI 457 4.07 4 0.998 180 4.1 4 1.08
technologies well for their career
Teachers can already accurately identify a 457 3.02 3 1.56 180 2.72 2 1.62
student’s usage of generative AI technologies to
partially complete an assignment
Generative AI technologies such as ChatGPT can 455 3.19 3 1.25 180 2.93 3 1.4
provide guidance for coursework as effectively as
human teachers
Using generative AI technologies such as Chat- 455 3.29 3 1.25 180 3.56 4 1.31
GPT to complete assignments undermines the
value of a university education
I can ask questions to generative AI technolo- 454 3.51 4 1.2 180 3.97 4 1.06
gies such as ChatGPT that I would otherwise
not voice out to my teacher. /Students can ask
questions to generative AI technologies such as
ChatGPT that they would otherwise not voice
out to their teacher
Generative AI technologies such as ChatGPT 452 3.66 4 1.15 180 4 4 1.17
will not judge me, so I feel comfortable with
it. /Students will not feel judged by generative
AI technologies such as ChatGPT, so they feel
comfortable with it
Generative AI technologies such as ChatGPT will 454 3.24 3 1.32 180 3.69 4 1.3
limit my opportunities to interact with others and
socialize while completing coursework. /Genera-
tive AI technologies such as ChatGPT will limit
students’ opportunities to interact with others
and socialize while completing coursework
Generative AI technologies such as ChatGPT will 454 3.3 3 1.33 180 3.74 4 1.41
hinder my development of generic or transfer-
able skills such as teamwork, problem-solving,
and leadership skills. /Generative AI technologies
such as ChatGPT will hinder students’ develop-
ment of generic or transferable skills such as
teamwork, problem-solving, and leadership skills
Table 2 (continued)
Item Students Teachers
N Mean Median SD N Mean Median SD
If a fully online programme with the assistance of 454 2.92 3 1.46 180 3.21 3 1.52
a personalized AI tutor was available, I would be
willing to pursue my degree through this option.
/If a fully online programme with the assistance
of a personalized AI tutor was available, students
should be open to pursuing their degree through
this option
I can become over-reliant on generative AI tech- 454 3.11 3 1.35 180 4.24 4 0.955
nologies. /Students can become over-reliant on
generative AI technologies
I believe generative AI technologies such as 454 3.8 4 1.06 180 3.83 4 1.12
ChatGPT can improve my digital competence. /I
believe Generative AI technologies such as Chat-
GPT can improve students’ digital competence
I believe generative AI technologies such as 455 3.67 4 1.18 180 3.63 4 1.36
ChatGPT can improve my overall academic per-
formance. /I believe Generative AI technologies
such as ChatGPT can improve students’ overall
academic performance
I believe generative AI technologies such as Chat- 453 4.23 4 0.848 180 4.06 4 1.01
GPT can help me save time. /I believe Genera-
tive AI technologies such as ChatGPT can help
students save time
I think generative AI technologies such as Chat- 455 3.46 4 1.27 180 3.31 3 1.45
GPT can help me become a better writer. /I think
Generative AI technologies such as ChatGPT can
help students become a better writer
I believe AI technologies such as ChatGPT can 455 3.84 4 1.13 180 3.77 4 1.26
provide me with unique insights and perspec-
tives that I may not have thought of myself. /I
believe AI technologies such as ChatGPT can
provide students with unique insights and
perspectives that they may not have thought of
themselves
I think AI technologies such as ChatGPT can 455 3.75 4 1.14 180 3.86 4 1.34
provide me with personalized and immediate
feedback and suggestions for my assignments. /
AI technologies such as ChatGPT can provide stu-
dents with personalized and immediate feedback
and suggestions for their assignments
I think AI technologies such as ChatGPT is a great 455 4.16 4 0.893 180 3.81 4 1.17
tool as it is available 24/7. /I think AI technologies
such as ChatGPT is a great tool for students as it
is available 24/7
I think AI technologies such as ChatGPT is a 455 3.91 4 1.12 180 3.77 4 1.29
great tool for student support services due to
anonymity
The respondents acknowledged the importance of learning to use generative AI tech-

nologies well for their careers (students: mean = 4.07, SD = 0.998; teachers: mean = 4.1,
SD = 1.08). However, both groups expressed doubt about teachers’ ability to accurately
identify a student’s usage of generative AI technologies for completing assignments (stu-
dents: mean = 3.02, SD = 1.56; teachers: mean = 2.72, SD = 1.62).
The responses to the remaining questions suggest that students and teachers recog-
nize potential benefits and drawbacks of AI technologies, including providing guidance
and personalized feedback, improving digital competence and academic performance,
and offering anonymity in student support services. However, there were concerns about
over-reliance on AI technologies, limited social interaction, and the potential hindrance
to the development of generic skills.
These findings highlight the need for a comprehensive AI policy in higher educa-
tion that addresses the potential risks and opportunities associated with generative AI
technologies. Based on these findings, some implications and suggestions for university
teaching and learning AI policy include:
1. Training: Providing training for both students and teachers on effectively using and
integrating generative AI technologies into teaching and learning practices.
2. Ethical Use and Risk Management: Developing policies and guidelines for ethical use
and risk management associated with generative AI technologies.
3. Incorporating AI without replacing human: Incorporating AI technologies as supple-
mentary tools to assist teachers and students, rather than as replacements for human
interaction.
4. Continuously Enhancing Holistic Competencies: Encouraging the use of AI technol-
ogies to enhance specific skills, such as digital competence and time management,
while ensuring that students continue to develop vital transferable skills.
5. Fostering a transparent AI environment: Fostering a transparent environment where
students and teachers can openly discuss the benefits and concerns associated with
using AI technologies in higher education.
6. Data Privacy and security: Ensuring data privacy and security while using AI tech-
nologies.
Overall, the survey results indicate an openness to adopting generative AI technolo-

gies in higher education and a recognition of the potential advantages and challenges.
Addressing these issues through informed policy and institutional support will be cru-
cial for maximizing the benefits of AI technologies in university teaching and learning.
Findings from the qualitative data

The qualitative data collected from students, teachers, and staff yielded valuable and rich
suggestions and comments. There are 10 main themes and 25 subthemes that emerged
from the qualitative data as presented in Table 3. From the data, we identified ten key
areas (i.e., the main themes) that are directly relevant to the planning of an AI policy for
teaching and learning in universities. These areas align well with the quantitative data
and are as follows:
(1) Understanding, identifying and preventing academic misconduct and ethical dilem-
mas
To address academic misconduct, universities must develop clear guidelines and strat-
egies for detecting and preventing the misuse of generative AI. Teachers emphasize the
importance of creating university-wide policies on how to test students suspected of
using AI to complete tasks in which AI use is prohibited or misused. As one student
stated, “A clear set of rules about what happens if AI is used and resources on informing
Table 3 Main themes and subthemes of qualitative data

Main themes Subthemes
1. Understanding, identifying and preventing aca- Develop guidelines and strategies for detecting and
demic misconduct and ethical dilemmas preventing the misuse of generative AI
Identify ethical dilemmas
Familiarize students with ethical issues
2. Addressing governance of AI: Data privacy, transpar- Be transparent about decisions concerning AI use
ency, accountability and security Ensure data privacy and security
Address ethical issues such as bias and stereotypes
3. Monitoring and evaluating AI implementation Conduct longitudinal experiments to examine the
effects of AI use
Collect feedback from teachers and students to make
informed decisions
4. Ensuring equity in access to AI technologies Provide resources and support to all students and staff
Ensure all students have access and training to AI tools
5. Attributing AI technologies Promote academic integrity in AI use
Develop guidelines on how to attribute generative AI’s
contribution to student work
6. Providing training and support for teachers, staff Enhance staff confidence and competence through
and students in AI literacy adequate training
Teach students how to use and critique the use of AI
technologies
Provide education on ethics; knowledge of the affor-
dances, use, and limitations; and capability to evaluate
AI outputs
7. Rethinking assessments and examinations Design assessments that integrate AI technologies to
enhance learning outcomes
Develop assessment strategies that focus on students’
critical thinking and analysis
8. Encouraging a balanced approach to AI adoption Recognize the potential benefits and limitations of
generative AI technologies
Avoid over-reliance on AI technologies
Use AI technologies as complementary tools
9. Preparing students for the AI-driven workplace Teach students how to use AI responsibly
Develop curricula that equip students with AI skills and
knowledge
Familiarize students with AI tools they will encounter for
university studies and future workplace
10. Developing student holistic competencies/generic Enhance students’ critical thinking to help them use AI
skills technologies effectively
Provide opportunities for developing competencies that
are impeded by AI use such as teamwork and leadership
students about the rule set are needed.” They also suggested, “Clearly stipulate in which
areas generative AI technologies are allowed and which are not. What are the procedures
to handle suspended cases? What are the consequences?” Another student mentioned
that “the level of restriction should be clarified.” Both teachers and students have also
suggested the use of assessments that minimize opportunities for AI misuse, such as oral
examinations or controlled settings where internet access is limited, to help maintain
academic integrity. Both teachers and students have also questioned “what is the defini-
tion of cheating?” in this AI era.
Teachers highlight the importance of identifying ethical dilemmas and recommend
familiarizing students with ethical issues, such as the boundaries between plagiarism
and inspiration and appropriate situations for seeking help from AI. Establishing clear
policies around AI use, including ethical guidelines and legal responsibilities, will help
students and staff navigate these complex issues. One teacher noted, “The education on
academic and research ethics should be strengthened.” Explicitly stipulating the areas
where AI is allowed and the procedures for handling suspected cases of misuse will help
maintain a transparent and equitable learning environment.
(2) Addressing governance of AI: data privacy, transparency, accountability and security
Universities must take responsibility for decisions made regarding the use of gen-
erative AI in teaching and learning, which includes being transparent about data
collection and usage, and being receptive to feedback and criticism. By disclosing
information about the implementation of generative AI, including the algorithms
employed, their functions, and any potential biases or limitations, universities can
foster trust and confidence among students and staff in AI technology usage. Teach-
ers emphasize the importance of addressing ethical concerns, privacy, security, and
other related issues when using generative AI technologies. Teachers commented
“In general, its impact is inevitable. It may negatively affect social consciousness and
responsibility. Depending on climate change management and its consequences, it may
contribute to the demise of a significant portion of humanity. It may also protect and
advance the interests of those who benefit from chaos.”
Privacy and Security: AI technologies rely on vast amounts of data, which raises
concerns about privacy and security if the data is not adequately protected. “Institu-
tions should ensure that the data used by generative AI technologies is kept private and
secure. This includes ensuring that any data used in training or testing the technology
is de-identified, and that appropriate security measures are in place to prevent unau-
thorized access or use of data.”
Transparency and Accountability: Universities should be transparent about the
use of generative AI in teaching and learning, which includes disclosing information
about the algorithms and their functions, as well as any potential biases or limitations
of the AI tools. “It is essential to recognize ethical dilemmas and consider privacy,
security, and related issues when employing generative AI technologies.”
The complexity of AI technologies can make it difficult to hold organizations and
individuals accountable for their decisions and actions. Institutions should address
ethical issues, such as potential discrimination, bias, and stereotypes, while ensuring
data privacy and security.
(3) Monitoring and evaluating AI implementation
To ensure the success of AI integration in university teaching and learning, con-

tinuous monitoring and evaluation of its implementation are necessary. Teachers rec-
ommend conducting longitudinal experiments in different areas to better understand
how AI affects students’ learning processes and outcomes. Regular assessments of
AI’s impact on teaching practices and student performance will help identify areas for
improvement and ensure that the technology is being used effectively and ethically.
One student mentioned, “The plan should include more experiments on conducting
the AI technologies on teaching.” By regularly collecting feedback from both teach-
ers and students, universities can make informed decisions about how to improve AI
implementation. Evaluating the effectiveness of AI tools in enhancing learning out-

comes is vital in determining their value and making adjustments as needed.
(4) Ensuring equity in access to AI technologies
Ensuring equitable access to AI technologies is crucial for fostering an inclusive

learning environment. Universities should work to provide resources and support
to all students and staff, regardless of their background or access to technology. This
may involve the procurement of AI tools, including AI detectors, for use by the entire
university community. By promoting equal access to AI technologies, universities can
help level the playing field and ensure that all students and staff have the opportunity
to benefit from the advantages offered by AI integration.
Equal access to AI technologies is essential for maintaining fairness in the educa-
tional environment. One teacher commented, “Same as all other resources, to incor-
porate this into current industries (especially education), fairness should be a top
priority. If the usage involves any kind of competition, e.g. access to ChatGPT should be
equal for all involved parties.” Another student highlights “Ethical dilemma includes
ensuring that the technology is not used to discriminate against individuals or groups,
and that it does not reinforce bias or stereotypes.” Universities should consider how to
ensure that all students have access and training to AI tools and resources, regardless
of their socio-economic backgrounds, in order to level the playing field and promote
inclusivity.
(5) Attributing AI technologies
Attribution is an important aspect of AI policy in university teaching and learn-

ing. One student remarked, “They are welcome to use AI for academic purposes while
requiring students to state clearly which part was helped by AI. This is similar to the
references and citation of current academic practice.” By requiring students to attrib-
ute AI-generated content, universities can promote academic integrity and ensure
that AI technologies are used ethically in the learning process. Furthermore, there
is a need for guidelines on how to fairly attribute generative AI’s contribution to stu-
dent work. “Ethics of use, knowledge of affordances, effective use, critique/evaluation of
outputs, and role/integration in workflows/product in study and professional settings”
may be included in the attribution.
(6) Providing training and support for teachers, staff and students in AI literacy
To ensure successful integration of AI in teaching and learning, universities must

provide adequate training and support for teachers, staff, and students. Teachers
express concerns about coping with this new trend, helping students use AI effec-
tively, and learning from student usage. As one teacher puts it, “Staff and students
need an educative approach to its ethical use.” Investing in training and resources
can help educators feel more confident and capable in navigating the complexities of
AI in their classrooms. This is supported by many students and teachers who believe
that institutions should provide training to faculty and staff on the appropriate use
of generative AI technologies in teaching and learning. “This training should include

information on selecting appropriate technologies, using them effectively, and manag-
ing the risks associated with their use,” they say. In addition, “teaching students how
to use the technology and how to critique it is probably central to successfully plan-
ning for the integration of AI in education.” Students suggest that “teaching students
the potential of using generative AI properly and critically can benefit from students
using AI hiddenly” and “relying on tools able to detect the use of a generative language
model, while being aware of the limits of such tools.”
AI literacy is crucial for both students and staff as they navigate the use of genera-
tive AI in teaching and learning. Teachers emphasize the need for education on eth-
ics, knowledge of AI tool affordances, effective use (e.g., prompt engineering), critique
and evaluation of outputs, and the role of AI in study and professional settings. A com-
prehensive AI literacy programme will help students and staff better understand and
responsibly utilize AI technologies in their academic and professional lives. By providing
training and resources on AI technologies, universities can empower students and staff
to make informed decisions about their use and potential applications in teaching and
learning.
(7) Rethinking assessments and examinations
The integration of generative AI in education calls for a re-evaluation of assessments

and examinations. Teachers suggest designing assessments that allow AI technologies
to enhance learning outcomes, rather than solely producing outputs. For example, one
teacher recommends “Promote assessments and activities where students can by them-
selves discover the limits of such techniques—and relativize the idea that they could be
useful to ‘cheat’.” This shift may necessitate the development of new assessment methods
that balance the benefits of AI with the need to maintain academic integrity. A student
stated, “Change of assessment methods to measure the true ‘understanding’ of students
instead of the ability to collect information (which can easily be done with AI tools).”
Universities may need to develop new assessment strategies that focus on students’
understanding, critical thinking, and analysis to prevent AI-generated content from
compromising the assessment process. A teacher noted that “… it is hard to assess most
of them, so we fall back on regurgitation”, a change is necessary.
(8) Encouraging a balanced approach to AI adoption
A balanced approach to AI adoption in university teaching and learning involves rec-

ognizing both the potential benefits and limitations of generative AI technologies. One
teacher suggests, “Be positive about this technological evolution and incorporate it to
develop new assignments and assessment.” This approach requires flexibility, striking a
balance between embracing new technology for its potential to enhance efficiency and
productivity while maintaining a focus on critical thinking and ethical considerations.
It is also important to encourage a balanced approach to AI adoption to avoid over-
reliance on these technologies. “We should learn how AI can assist us, but not replace
schoolwork,” one teacher advised. This approach involves using AI technologies as
complementary tools to support learning rather than relying on them as a substitute for
traditional teaching methods. Students should be encouraged to use AI as an aid to their
learning process and not solely depend on it for academic success.
(9) Preparing students for the AI-driven workplace
Preparing students for an AI-driven workplace involves teaching them how to use
AI responsibly, ethically, and effectively. Universities should develop curricula that
reflect the increasing prominence of AI in various industries, ensuring that students
are equipped with the skills and knowledge to navigate the evolving workplace land-
scape. This includes teaching students how to integrate AI into their workflows,
evaluate the effectiveness of AI tools, and understand their role in professional set-
tings. As one teacher notes, “Teaching students how to use it properly and under-
standing its limitations and strengths would be useful.”
Integrating AI technologies into teaching and learning involves familiarizing stu-
dents with AI tools they will likely encounter during their university studies and in
the workplace, as mentioned by a student who said, “Teach students how to best use
AI tools and make AI tools a common part of education, just like PowerPoint and
Excel.” Teachers suggest guiding students to recognize ethical issues and helping
them self-appropriate AI in study and work settings.
As the workplace increasingly adopts AI technologies, universities should prepare
students for this shift. One student stated, “Plans should be implemented to assist
students in making better and more constructive use of AI in learning, career plan-
ning, and personal development.”
(10) Developing student holistic competencies/generic skills
Teachers have highlighted the importance of enhancing critical thinking, digital

literacy, information literacy, and professional ethics among students to help them
make effective and ethical use of AI technologies. To harness the potential of gen-
erative AI technologies, teachers advocate for an emphasis on teaching students to
assess the reliability of content, understand biases, and evaluate the accuracy and
relevance of AI-generated information. One teacher suggests that “Enhancement on
critical thinking among students is definitely a must, in order to make good use of
such AI technologies.” Another teacher emphasizes the importance of “Teaching stu-
dents how to use the technology, and how to critique it, is probably central to success-
fully planning for the integration of AI in education.”
To successfully embrace generative AI technologies, universities should prioritize
fostering critical thinking among students. One student suggested, “Plans on how to
maintain students’ interest and motivation to engage in deep and critical thinking,
diversify perspectives and expand horizons.”
Developing holistic competencies and generic skills in students is an essential goal
of education. “However there are many things, even far more important things, to
education…: the education of character; rhetoric and analytical skills; public speak-
ing; creativity; memorisation; all embodied skills,” a teacher argued. Incorporating
AI technologies into teaching and learning may hinder students’ development of
competencies such as teamwork, leadership, empathy, and creativity skills. There-

fore, universities need to continuously find opportunities for students to develop
these skills, preparing them for the AI-driven workplace where they need to be
adaptable, resilient, and transformational.
Discussion
Triangulating quantitative and qualitative data
The quantitative findings support the key areas found in the qualitative data for AI
integration in education. The quantitative data reveals that both students and teach-
ers share concerns about the potential misuse of AI technologies, such as ChatGPT,
in assignments (students: mean 3.67, teachers: mean 3.93). This emphasizes the need
for guidelines and strategies to prevent academic misconduct. Furthermore, there is
significant agreement among students and teachers on the necessity for higher educa-
tion institutions to implement a plan for managing the potential risks associated with
using generative AI technologies (students: mean 4.5, teachers: mean 4.54), highlight-
ing the importance of addressing data privacy, transparency, accountability, and secu-
rity. The overall positive perception of AI technologies integration within education
implies that proper policies should be in place to ensure responsible AI incorporation
in higher education.
The concern that some students might use generative AI technologies to gain an
advantage in their assignments (students: mean 3.67, teachers: mean 3.93) under-
scores the importance of ensuring equal access to AI technologies for all students.
Moreover, the consensus that students must become proficient in using generative AI
technologies for their careers (students: mean 4.07, teachers: mean 4.1) highlights the
need for AI literacy and training for all stakeholders in the educational process, pre-
paring students for the AI-driven workplace.
Interestingly, teachers and students are unsure if teachers can accurately identify a
student’s use of generative AI technologies to partially complete an assignment (stu-
dents: mean 3.02, teachers: mean 2.72), yet they also believe that AI technologies can
provide unique insights and perspectives and personalized feedback. This suggests
that rethinking assessment methods may be necessary.
Data indicating that neither students nor teachers believe AI technologies will
replace teachers in the future (students: mean 2.14, teachers: mean 2.26) supports the
need for a balanced approach to AI adoption, utilizing AI technologies as comple-
mentary tools rather than substitutes for traditional teaching methods. Finally, con-
cerns that generative AI technologies could hinder students’ development of generic
or transferable skills, such as teamwork, problem-solving, and leadership (students:
mean 3.3, teachers: mean 3.74), emphasize the importance of focusing on students’
holistic competencies and generic skills in preparation for the AI-driven workplace.
Key areas versus UNESCO’s recommendations on AI education policy

The original plan for the study was to use UNESCO’s recommendations as a basis
for developing AI education policy framework for university teaching and learning
through inputs from various stakeholders to identify any gaps in the framework and
modified accordingly. Although the recommendations from UNESCO can provide a

high-level guideline for this study, it was clear that there are several key differences
between UNESCO recommendations and the ten key areas that were identified for
integrating AI in university teaching and learning.
As the UNESCO’s AI and Education: Guidance for Policy-Makers was written
before the availability of GPT 3.5 and 4, the recommendations would not have fully
addressed the current opportunities and threats of the advances in the GPT technolo-
gies for education.
Moreover, the UNESCO recommendations are intended for education in general and
do not specifically cater to the needs of university teaching and learning. The UNESCO
recommendations are high-level and general aimed at helping policy-makers better
understand the possibilities and implications of AI for teaching and learning to help
achieve Sustainable Development Goal 4: Ensure inclusive and equitable quality educa-
tion and promote lifelong learning opportunities for all (UNESCO, 2021a), while the ten
key areas are more specific, practical, and tailored to university teaching and learning.
Furthermore, most of the existing AI education policies tend to emphasize the instru-
mental role of AI for workforce development (Schiff, 2022), but the key areas based on
the findings of this study also address the transformative potential and ethical considera-
tions relating to AI use in higher education in addition to its instrumental purposes.
The ten key areas were developed based on direct input from stakeholders who have
vested interests in university teaching and learning, which makes them more relevant
and grounded in practice. For example, UNESCO’s recommendation to develop a mas-
ter plan for using AI for education management, teaching, learning, and assessment
emphasizes the need for a comprehensive and strategic approach to integrating AI in
various aspects of education. This includes not only teaching and learning but also the
broader aspects of education management, such as administration, resource allocation,
and policy development. The focus here is on creating an overarching framework that
guides the implementation of AI in education as a whole. On the other hand, the key
area derived from the qualitative findings of rethinking assessments and examinations
delves deeper into a specific aspect of education: the evaluation of students’ learning.
This area acknowledges that the integration of generative AI in education necessitates a
re-evaluation of traditional assessment methods. The focus here is on designing assess-
ments that allow AI technologies to enhance learning outcomes while maintaining
academic integrity. This involves developing new assessment strategies that focus on stu-
dents’ understanding, critical thinking, and analysis rather than just their ability to col-
lect information. In short, UNESCO’s recommendations highlight the “what”, whereas
the ten key areas of this study detail the “how” of AI education policy.
In essence, the difference between these two areas lies in their scope and focus.
UNESCO’s recommendation is broader, encompassing various aspects of education
and advocating for a comprehensive master plan. Its primary target is policymakers. In
contrast, the key area on rethinking assessments and examinations is more specific and
user centred, catering for various higher education stakeholders including students and
teachers as well as addressing the challenges and opportunities associated with AI inte-
gration in university teaching and learning.
AI Ecological Education Policy Framework

a. Understanding, identifying, and
preventing academic misconduct and
Governance ethical dilemmas
Dimension b. Addressing governance of AI: data
privacy, transparency, accountability,
[Senior and security
c. Atributing AI technologies
Management] d. Ensuring equity in access to AI
Students,
Teachers, Staff,
Management,
External
Agents Operational
Pedagogical Dimension
Dimension [Teaching and
[Teachers] Learning and IT
staff]
a. Monitoring and evaluating AI

a. Rethinking assessments and
implementation
examinations
b. Providing training and support
b. Developing student holistic
for teachers, staff, and
competencies/generic skills
students in AI literacy
c. Preparing students for the AI-driven
workplace
d. Encouraging a balanced approach to AI
adoption
Fig. 1 AI ecological education policy framework
AI ecological education policy framework

In order to turn policy recommendations into action plans, the ten key areas have been
further organised into three dimensions—Pedagogical, Ethical, and Operational into the
AI Ecological Education Policy Framework. Each dimension is led by a responsible party
(see Fig. 1). This framework allows for a more nuanced understanding of the multifac-
eted implications of AI integration in university settings and ensures that stakeholders
consider the broader context of AI adoption and its impact on various aspects of teach-
ing and learning.
Pedagogical dimension (teachers)

This dimension focuses on the teaching and learning aspects of AI integration. It
includes the following key areas:
a. Rethinking assessments and examinations

b. Developing student holistic competencies/generic skills
c. Preparing students for the AI-driven workplace
d. Encouraging a balanced approach to AI adoption
Governance dimension (senior management)

This dimension emphasizes the governance considerations surrounding AI usage in edu-
cation. It encompasses the following key areas:
a. Understanding, identifying, and preventing academic misconduct and ethical dilem-

mas
b. Addressing governance of AI: data privacy, transparency, accountability, and security
c. Attributing AI technologies
d. Ensuring equity in access to AI technologies
Operational dimension (teaching and learning and IT staff)

This dimension concentrates on the practical implementation of AI in university set-
tings. It includes the following key areas:
a. Monitoring and evaluating AI implementation

b. Providing training and support for teachers, staff, and students in AI literacy
For the Pedagogical dimension, the framework emphasizes the need to adapt teaching
methods and assessment strategies in response to AI’s growing capabilities, preparing
students for an increasingly AI-driven workplace. By focusing on pedagogy, the frame-
work ensures that AI technologies are harnessed to enhance learning outcomes and
develop critical thinking, creativity, and other essential skills, rather than undermining
academic integrity. Teachers are the initiator for the Pedagogical Dimension, as they are
the ones who design and implement lesson plans, activities, and assessments that utilize
AI technologies. They will need to have the expertise to determine how AI can best sup-
port and enhance students’ learning experiences. At the same time, they must possess
the necessary awareness and knowledge to educate students on the potential risks asso-
ciated with the use of generative AI in learning, particularly in assessments and assign-
ments, such as plagiarism and contract cheating. Teachers need to foster ethical use of
AI, for example through proper attribution to acknowledge the contributions of AI tech-
nologies in student work, and develop assessment tasks that require critical and analyti-
cal thinking to avoid AI-assisted plagiarism. By assigning teachers the responsibility for
this dimension, we ensure that AI tools are used in a way that is pedagogically sound and
enhances the learning outcomes of students.
The Governance dimension highlights the importance of addressing issues related to
academic misconduct, data privacy, transparency, and accountability. The framework
ensures that stakeholders understand and address the ethical challenges associated with
AI technologies, fostering responsible use and helping to maintain trust within the uni-
versity community. This focus on governance encourages universities to develop clear
policies and guidelines, ensuring that students and staff can navigate the complex ethical
landscape surrounding AI.
Senior management will be the initiator for the Governance Dimension of the AI Ecol-
ogy Framework. As they hold decision-making authority, they are tasked with develop-
ing and enforcing policies, guidelines, and procedures that address the ethical concerns
surrounding AI use in education. These include academic integrity, data privacy, trans-
parency, accountability, and security. Senior management’s role is to ensure that AI is
used responsibly and ethically, fostering a learning environment that is fair, equitable,
and inclusive.
The Operational dimension of the framework underscores the need for ongoing moni-
toring, evaluation, and support to ensure the effective and equitable implementation of
AI technologies. By considering operational aspects, the framework encourages univer-
sities to provide training, resources, and support to all stakeholders, promoting equal
access to AI technologies and fostering an inclusive learning environment. Furthermore,
the operational dimension emphasizes the importance of continuous improvement and
adaptation, enabling universities to refine their AI integration strategies in response to
new insights and changing needs.
Teaching and Learning and IT staff will be tasked to look after the Operational Dimen-
sion. They play a crucial role in managing and maintaining the AI technologies used in
the educational setting. Their tasks include providing training and support for both stu-
dents and staff, ensuring the proper functioning of AI tools, and addressing any techni-
cal issues that may arise. They can ensure that AI technologies are seamlessly integrated
into the educational environment, minimizing disruptions and maximizing their poten-
tial benefits.
It is crucial to recognize that the responsibility of each dimension in the ecological
framework should not be viewed in isolation. Collaboration and communication among
all stakeholders (universities, teachers, students, staff and external agents such as accred-
itation, quality assurance bodies) are essential to ensure the successful implementation
of any policy. Each group should actively participate in the development and execution
of AI-related initiatives and work together to achieve the desired outcomes in university
teaching and learning.
Conclusions
This study aims to establish an AI education policy for university teaching and learn-
ing, addressing concerns related to the use of text-generating AI in academic environ-
ments, such as cheating and plagiarism. The findings of this study yield ten key areas in
AI education policy planning, from which an AI Ecological Education Policy Framework
is constructed to fulfil the objective of the study. However, this study has some limita-
tions, including a relatively small sample size that may not be representative of all edu-
cational institutions. Additionally, the research only focused on text-based generative AI
technology and did not explore other types or variations. Lastly, the study relied on self-
reported data from participants, which may be subject to bias or inaccuracies.
This study proposes an AI Ecological Education Policy Framework to address the
diverse implications of AI integration in university settings. The framework consists of
three dimensions—Pedagogical, Governance, and Operational—each led by a respon-
sible party. This structure allows for a more comprehensive understanding of AI inte-
gration implications in teaching and learning settings and ensures stakeholders are
aware of their responsibilities. By adopting this framework, educational institutions can
align actions with their policy, ensuring responsible and ethical AI usage while maxi-
mizing potential benefits. However, more research is necessary to fully comprehend the
potential advantages and risks associated with AI in academic settings. Merely advocat-
ing for AI implementation in education is insufficient; stakeholders need to carefully
evaluate which AI technologies to employ, determine the best methods for their use, and
understand their true capabilities.
Acknowledgements
The author wishes to thank the students and teachers who participated the survey.
Author contributions
The study is conducted solely by the author.
Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable
request.
Declarations
Competing interests
The author declares that one has no competing interests.
Received: 30 March 2023 Accepted: 7 June 2023
References
Abduljabbar, R., Dia, H., Liyanage, S., & Bagloee, S. A. (2019). Applications of artificial intelligence in transport: An overview.
Sustainability, 11(1), 189. https://doi.org/10.3390/su11010189
Adiguzel, T., Kaya, M. H., & Cansu, F. K. (2023). Revolutionizing education with AI: Exploring the transformative potential of
ChatGPT. Contemporary Educational Technology, 15(3), ep429. https://doi.org/10.30935/cedtech/13152
AI regulation (2023). Legal and Ethical Aspects of ChatGPT: EU Parliament’s Amendment, French Experts’ Opinion on Ethical Issues
and Other Useful Resources. Retrieved from https://ai-regulation.com/legal-and-ethical-aspects-of-chatgpt/
Al Braiki, B., Harous, S., Zaki, N., & Alnajjar, F. (2020). Artificial intelligence in education and assessment methods. Bulleting of
Electrical Engineering and Informatics, 9(5), 1998–2007. https://doi.org/10.11591/eei.v9i5.1984
Aoun, J. E. (2017). Robot-proof: Higher education in the age of artificial intelligence. The MIT Press.
Atlas, S. (2023). ChatGPT for higher education and professional development: A guide to conversational AI. https://digitalcom
mons.uri.edu/cba_facpubs/548
Bholat, D., & Susskind, D. (2021). The assessment: Artificial intelligence and financial services. Oxford Review of Economic Policy,
37(3), 417–434. https://doi.org/10.1093/oxrep/grab015
Buckley, R. P., Zetzsche, D. A., Arner, D. W., & Tang, B. W. (2021). Regulating artificial intelligence in finance: Putting the human in
the loop. The Sydney Law Review, 43(1), 43–81.
Cassidy, C. (2023). Universities to return to pen and paper exams after students caught using AI to write essays. The Guardian.
https://www.theguardian.com/australia-news/2023/jan/10/universities-to-return-to-pen-and-paper-exams-after-stude
nts-caught-using-ai-to-write-essays
Cavendish, C. (2023). ChatGPT will force school exams out of the dark ages. Financial Times. https://www.ft.com/content/41243
091-d8d7-4b74-9ad1-5341c16c869f
Chan, C.K.Y. (2023). Is AI changing the rules of academic misconduct? An in-depth look at students’ perceptions of ’AI-giarism’.
https://arxiv.org/abs/2306.03358
Chan, C. K. Y., & Hu, W. (2023). Students’ voices on generative AI: Perceptions, benefits, and challenges in higher education.
https://doi.org/10.48550/arXiv.2305.00290
Chan, CKY & Chen, S. (2023). Student Partnership in Assessment in Higher Education: A Systematic Review. https://doi.org/10.
1080/02602938.2023.2224948
Chan, C. K. Y., & Lee, K. K. W. (2023). The AI generation gap: Are Gen Z students more interested in adopting generative AI such
as ChatGPT in teaching and learning than their Gen X and Millennial Generation teachers? arxiv:2305.02878
Chan, C. K. Y., & Tsi, L. H. Y. (2023). The AI Revolution in Education: Will AI Replace or Assist Teachers in Higher Education?
[Preprint]. arxiv:2305.01185
Chan, C. K. Y., & Zhou, W. (2023). Deconstructing Student Perceptions of Generative AI (GenAI) through an Expectancy Value
Theory (EVT)-based Instrument [Preprint]. arxiv:2305.01186
Chatterjee, S. (2020). AI strategy of India: Policy framework, adoption challenges and actions for government. Transforming
Government, 14(5), 757–775. https://doi.org/10.1108/TG-05-2019-0031
Civil, B. (2023, March 16). ChatGPT can hinder students’ critical thinking skills: Artificial intelligence is changing how students
learn to write. The Queen’s Journal. https://www.queensjournal.ca/story/2023-03-16/opinions/chatgpt-can-hinder-stude
nts-critical-thinking-skills/
Cotton, D. R. E., Cotton, P. A., & Shipway, J. R. (2023). Chatting and cheating: Ensuring academic integrity in the era of ChatGPT.
Innovations in Education and Teaching International. https://doi.org/10.1080/14703297.2023.2190148
Dexe, J., & Franke, U. (2020). Nordic lights? National AI policies for doing well by doing good. Journal of Cyber Policy, 5(3),
332–349. https://doi.org/10.1080/23738871.2020.1856160
Eggmann, F., Weiger, R., Zitzmann, N. U., & Blatz, M. B. (2023). Implications of large language models such as ChatGPT for
dental medicine. Journal of Esthetic and Restorative Dentistry. https://doi.org/10.1111/jerd.13046
Federspiel, F., Mitchell, R., Asokan, A., Umana, C., & McCoy, D. (2023). Threats by artificial intelligence to human health and
human existence. BMJ Global Health, 8(5), e010435. https://doi.org/10.1136/bmjgh-2022-010435
Feldstein, S. (2019). The road to digital unfreedom: How artificial intelligence is reshaping repression. Journal of Democracy,
30(1), 40–52. https://doi.org/10.1353/jod.2019.0003
Floridi, L. (2021). A Unified Framework of Five Principles for AI in Society. In Ethics, Governance, and Policies in Artificial Intel-
ligence (Vol. 144, pp. 5–17). Springer International Publishing AG. https://doi.org/10.1007/978-3-030-81907-1_2
Gellai, D. B. (2022). Enterprising academics: Heterarchical policy networks for artificial intelligence in British higher education.
ECNU Review of Education. https://doi.org/10.1177/20965311221143798
Greiman, V. A. (2021). Human rights and artificial intelligence: A universal challenge. Journal of Information Warfare, 20(1),
50–62.
Hogenhout, L. (2021). A Framework for Ethical AI at the United Nations. https://doi.org/10.48550/arxiv.2104.12547
IMDA & PDPC (2020). Model Artificial Intelligence Governance Framework. Retrieved from https://www.pdpc.gov.sg/-/media/
files/pdpc/pdf-files/resource-for-organisation/ai/sgmodelaigovframework2.pdf
Intelligent.com. (2023, January 23). Nearly 1 in 3 College Students Have Used ChatGPT on Written Assignments. https://www.intel
ligent.com/nearly-1-in-3-college-students-have-used-chatgpt-on-written-assignments/
Kasneci, E., Seßler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., & Kasneci, G. (2023). ChatGPT for
good? On opportunities and challenges of large language models for education. Learning and Individual Differences, 103,
102274.
Korn, J. & Kelly, S. (2023, January 5). New York City public schools ban access to AI tool that could help students cheat. CNN. https://
edition.cnn.com/2023/01/05/tech/chatgpt-nyc-school-ban/index.html
Luan, H., Geczy, P., Lai, H., Gobert, J., Yang, S. J. H., Ogata, H., Baltes, J., Guerra, R., Li, P., & Tsai, C.-C. (2020). Challenges and future
directions of big data and artificial intelligence in education. Frontiers in Psychology, 11, 580820. https://doi.org/10.3389/
fpsyg.2020.580820
Luckin, R. (2017). Towards artificial intelligence-based assessment systems. Nature Human Behaviour, 1, 0028. https://doi.org/
10.1038/s41562-016-0028
McKinsey Consultant. (2023). What is generative AI? [Article]. Retrieved February 12, 2023, from https://www.mckinsey.com/
featured-insights/mckinsey-explainers/what-is-generative-ai
Ocaña-Fernández, Y., Valenzuela- Fernández, L. A., & Garro-Aburto, L. L. (2019). Artificial intelligence and its implications in
higher education. Journal of Educational Psychology, 7(2), 553–568. https://doi.org/10.20511/pyr2019.v7n2.274
Oliver, J. (2023). John Oliver on new AI programs: ‘The potential and the peril here are huge’. The Guardian. https://www.thegu
ardian.com/tv-and-radio/2023/feb/27/john-oliver-new-ai-programs-potential-peril
Pelletier, K., McCormack, M., Reeves, J., Robert, J., Arbino, N., Al-Freih, M., Dickson-Deane, C., Guevara, C., Koster, L., Sánchez-
Mendiola, M., Bessette, L. S., & Stine, J. (2022). EDUCAUSE Horizon Report, teaching and learnig edition. EDUCAUSE.
Popenici, S. A. D., & Kerr, S. (2017). Exploring the impact of artificial intelligence on teaching and learning in higher education.
Research and Practice in Technology Enhanced Learning, 12, 22. https://doi.org/10.1186/s41039-017-0062-8
Renda, A. (2020). Europe: Toward a Policy Framework for Trustworthy AI. In M. D. Dubber, F. Pasquale, & S. Das (Eds.), The Oxford
Handbook of Ethics of AI (pp. 650–666). Oxford Academic. https://doi.org/10.1093/oxfordhb/9780190067397.013.41
Sam, A. K., & Olbrich, P. (2023). The need for AI ethics in higher education. In C. C. Corrigan, S. A. Asakipaam, J. J. Kponyo, & C.
Luetge (Eds.), AI ethics in higher education: Insights from Africa and beyond (pp. 3–10). Springer. https://doi.org/10.1007/
978-3-031-23035-6_1
Schiff, D. (2022). Education for AI, not AI for Education: The Role of Education and Ethics in National AI Policy Strategies. Inter-
national Journal of Artificial Intelligence in Education, 32(3), 527–563. https://doi.org/10.1007/s40593-021-00270-2
Sinhaliz, S., Burdjaco, Z., & Du Preez, J. (2023). How ChatGPT Could Revolutionize Academia. IEEE Spectrum. https://spectrum.
ieee.org/how-chatgpt-could-revolutionize-academia
Southgate, E. (2020). Artificial intelligence, ethics, equity and higher education: A “beginning-of-the-discussion” paper. National
Centre for Student Equity in Higher Education, Curtin University, and the University of Newcastle. http://ncsehe.edu.au/
wp-content/uploads/2020/07/Southgate_AI-Equity-Higher-Education_FINAL.pdf
Swiecki, Z., Khosravi, H., Chen, G., Martinez-Maldonado, R., Lodge, J. M., Milligan, S., Selwyn, N., & Gašević, D. (2022). Assessment
in the age of artificial intelligence. Computers and Education Artificial Intelligence, 3, 100075. https://doi.org/10.1016/j.
caeai.2022.100075
Tanveer, M., Hassan, S., & Bhaumik, A. (2020). Academic policy regarding sustainability and artificial intelligence (AI). Sustain-
ability, 12(22), 9435. https://doi.org/10.3390/su12229435
TEQSA (February 28, 2023). Artificial Intelligence: advice for students. Retrieved from https://www.teqsa.gov.au/students/artif
icial-intelligence-advice-students
UNESCO. (2021a). AI and education: Guidance for policy-makers. UNESCO.
UNESCO. (2021b). Recommendations on the Ethics of Artificial Intelligence. UNESCO.
UNESCO (2023). Ethics of Artificial Intelligence. Retrieved from https://www.unesco.org/en/artifi cial-intelligence/recommenda
tion-ethics
Wang, S., Wang, G., Chen, X., Wang, W., & Ding, X. (2021). A review of content analysis on China artificial intelligence (AI) educa-
tion policies. In W. Wang, G. Wang, X. Ding, & B. Zhang (Eds.), Artificial intelligence in education and teaching assessment
(pp. 1–8). Springer. https://doi.org/10.1007/978-981-16-6502-8_1
Warschauer, M., Tseng, W., Yim, S., Webster, T., Jacob, S., Du, Q, & Tate, T. (2023). The affordances and contradictions of AI-
generated text for second language writers. https://doi.org/10.2139/ssrn.4404380
Wood, P. (2023, February 28). Oxford and Cambridge ban AI language tool GPT-3 over fears of plagiarism. inews.co.uk. https://
inews.co.uk/news/oxford-cambridge-ban-chatgpt-plagiarism-universities-2178391
World Economic Forum. (2023). Model Artificial Intelligence Governance Framework and Assessment Guide. Retrieved from
https://www.weforum.org/projects/model-ai-governance-framework
Wu, J., Wang, X., Dang, Y., & Lv, Z. (2022). Digital twins and artificial intelligence in transportation infrastructure: Classification,
application, and future research directions. Computers and Electrical Engineering, 101, 107983. https://doi.org/10.1016/j.
compeleceng.2022.107983
Yau, C., & Chan, K. (2023, February 17). University of Hong Kong temporarily bans students from using ChatGPT, other AI-based
tools in coursework. South China Morning Post. https://www.scmp.com/news/hong-kong/education/article/32106
50/university-hong-kong-temporarily-bans-students-using-chatgpt-other-ai-based-tools-coursework
Yu, K.-H., Beam, A. L., & Kohane, I. S. (2018). Artificial intelligence in healthcare. Nature Biomedical Engineering, 2, 719–731.
https://doi.org/10.1038/s41551-018-0305-z
Zhai, X. (2022). ChatGPT user experience: Implications for education. Available at SSRN 4312418.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

AI Policy Education ChatGPT Policy in HE

Uploaded by

Copyright:

Available Formats

AI Policy Education ChatGPT Policy in HE

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

AI Policy Education ChatGPT Policy in HE

Uploaded by

Copyright:

Available Formats

Chan ﻿Int J Educ Technol High Educ (2023) 20:38 International Journal of Educational

RESEARCH ARTICLE Open Access

A comprehensive AI policy education

• ProposedAI Ecological Education Policy Framework for university teaching and

Keywords: AI policy framework, Artificial intelligence, ChatGPT, Ethics, Assessment

Generative AI and generative pre‑trained transformers

Rationale for an artificial intelligence education policy

Existing policy on artificial intelligence

Table 1 Compilation of fundamental ethical principles (IMDA &PDPC, 2020)

Existing policy on AI in education

1. Interdisciplinary planning and inter-sectoral governance: This recommendation

Quantitative data (survey data) and descriptive analysis

Qualitative data (open‑ended data) and thematic analysis

Table 2 Descriptive analysis for quantitative results

The respondents acknowledged the importance of learning to use generative AI tech-

Overall, the survey results indicate an openness to adopting generative AI technolo-

Findings from the qualitative data

Table 3 Main themes and subthemes of qualitative data

(3) Monitoring and evaluating AI implementation

To ensure the success of AI integration in university teaching and learning, con-

implementation. Evaluating the effectiveness of AI tools in enhancing learning out-

(4) Ensuring equity in access to AI technologies

Ensuring equitable access to AI technologies is crucial for fostering an inclusive

(5) Attributing AI technologies

Attribution is an important aspect of AI policy in university teaching and learn-

To ensure successful integration of AI in teaching and learning, universities must

of generative AI technologies in teaching and learning. “This training should include

(7) Rethinking assessments and examinations

The integration of generative AI in education calls for a re-evaluation of assessments

(8) Encouraging a balanced approach to AI adoption

A balanced approach to AI adoption in university teaching and learning involves rec-

(9) Preparing students for the AI-driven workplace

(10) Developing student holistic competencies/generic skills

Teachers have highlighted the importance of enhancing critical thinking, digital

competencies such as teamwork, leadership, empathy, and creativity skills. There-

Key areas versus UNESCO’s recommendations on AI education policy

modified accordingly. Although the recommendations from UNESCO can provide a

AI Ecological Education Policy Framework

a. Monitoring and evaluating AI

Fig. 1 AI ecological education policy framework

AI ecological education policy framework

Pedagogical dimension (teachers)

a. Rethinking assessments and examinations

Governance dimension (senior management)

a. Understanding, identifying, and preventing academic misconduct and ethical dilem-

Operational dimension (teaching and learning and IT staff)

a. Monitoring and evaluating AI implementation

Availability of data and materials

Received: 30 March 2023 Accepted: 7 June 2023

You might also like

Chan Int J Educ Technol High Educ (2023) 20:38 International Journal of Educational