View in EDS HTML Full Text PDF Full Text

Features, Components and Processes of Developing Policy for Artificial Intelligence in Education (AIED): Toward a Sustainable AIED Development and Adoption

Saved in:
Bibliographic Details
Title: Features, Components and Processes of Developing Policy for Artificial Intelligence in Education (AIED): Toward a Sustainable AIED Development and Adoption
Language: English
Authors: Awol Endris, Ahmed Tlili, Ronghuai Huang, Lin Xu, TingWen Chang, Sanjaya Mishra
Source: Leadership and Policy in Schools. 2025 24(1):233-241.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 9
Publication Date: 2025
Document Type: Journal Articles
Reports - Descriptive
Descriptors: Policy Formation, Artificial Intelligence, Technology Uses in Education, Sustainable Development, Educational Policy, Educational Improvement, Ethics
DOI: 10.1080/15700763.2024.2312999
ISSN: 1570-0763
1744-5043
Abstract: Governments and private sectors are now putting in place the needed resources and infrastructure to harness the power of emerging technologies in education. One of these technologies is Artificial Intelligence (AI) which gained increasing attention due to its potential to enhance learning and teaching experiences, hence achieving better learning outcomes. However, AIED also comes with several concerns that raise continuous questions about its safe and effective adoption. The application of AIED needs to be planned and executed properly. One of the basic requirements for this to happen is that a comprehensive national policy on the use of AIED is put in place to guide its implementation and evaluate its effectiveness. Limited information exists in the literature on how to write an AIED policy. To address this research gap, this study therefore discusses the features, major components and processes that countries are advised to adopt for a comprehensive AIED policy development. Specifically, this study highlights four features for a good AIED policy, namely contextual, consultative, dynamic, and, implementable and measurable. It further proposes seven steps for an AIED policy development, namely (1) pre-drafting consultation, (2) stakeholder survey, (3) writing draft policy, (4) discussion on draft policy, (5) adoption and communication of policy document, (6) policy implementation plan, and (7) policy evaluation.
Abstractor: As Provided
Entry Date: 2025
Accession Number: EJ1459780
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwHMzESpGp9wed5Bbo1U05NlAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDLNXM_BJKHBtkS2eYAIBEICBmxpChAKCvoLNqoijmVseBbqt3cFfc8eCKtzzwv-GLRFCg-w3YDBsmY_PHxomhpqIc3L60bhxpSqyjQ86fz9KfF7a6CrmKV54T0v7gtyciNEt0rr7m1YpPSxWQ5TwwDJYvogqerEv8DDvxIU5h-4A5FnR-Q62DyMuXQ2EGG2ihFTt3Xhxl8L4vKmrV9BaKBNB6ehWE6Ana6boq2zq
Text:
  Availability: 1
  Value: <anid>AN0182583709;o8o01jan.25;2025Feb03.04:56;v2.2.500</anid> <title id="AN0182583709-1">Features, Components and Processes of Developing Policy for Artificial Intelligence in Education (AIED): Toward a Sustainable AIED Development and Adoption </title> <sbt id="AN0182583709-2">Introduction</sbt> <p>Governments and private sectors are now putting in place the needed resources and infrastructure to harness the power of emerging technologies in education. One of these technologies is Artificial Intelligence (AI) which gained increasing attention due to its potential to enhance learning and teaching experiences, hence achieving better learning outcomes. However, AIED also comes with several concerns that raise continuous questions about its safe and effective adoption. The application of AIED needs to be planned and executed properly. One of the basic requirements for this to happen is that a comprehensive national policy on the use of AIED is put in place to guide its implementation and evaluate its effectiveness. Limited information exists in the literature on how to write an AIED policy. To address this research gap, this study therefore discusses the features, major components and processes that countries are advised to adopt for a comprehensive AIED policy development. Specifically, this study highlights four features for a good AIED policy, namely contextual, consultative, dynamic, and, implementable and measurable. It further proposes seven steps for an AIED policy development, namely (<reflink idref="bib1" id="ref1">1</reflink>) pre-drafting consultation, (<reflink idref="bib2" id="ref2">2</reflink>) stakeholder survey, (<reflink idref="bib3" id="ref3">3</reflink>) writing draft policy, (<reflink idref="bib4" id="ref4">4</reflink>) discussion on draft policy, (<reflink idref="bib5" id="ref5">5</reflink>) adoption and communication of policy document, (<reflink idref="bib6" id="ref6">6</reflink>) policy implementation plan, and (<reflink idref="bib7" id="ref7">7</reflink>) policy evaluation.</p> <p>One of the six challenges pointed out by the United Nations Educational, Scientific and Cultural Organization (UNESCO, [<reflink idref="bib24" id="ref8">24</reflink>]) to achieve sustainable development of AIED is comprehensive public policy. Schiff ([<reflink idref="bib21" id="ref9">21</reflink>]) analyzed 24 national AI policies and found that the use of AI in education is largely absent from policy discussions. The approach has been to focus on education for AI rather than AI for education. Notwithstanding the potential benefits of using AI in education for teaching support, grading, feedback, content generation and recommendation, there is a growing concern about ethical challenges, lack of technological readiness and understanding of AI, replicability, transparency and data privacy (Yan et al., [<reflink idref="bib27" id="ref10">27</reflink>]). Miao et al. ([<reflink idref="bib16" id="ref11">16</reflink>]) emphasized that to leverage the opportunities and reduce potential risks, key policy questions must be addressed: "(<reflink idref="bib1" id="ref12">1</reflink>) How can AI be leveraged to enhance education? (<reflink idref="bib2" id="ref13">2</reflink>) How can we ensure the ethical, inclusive and equitable use of AI in education? (<reflink idref="bib3" id="ref14">3</reflink>) How can education prepare humans to live and work with AI?"</p> <p>Policy is an intent made by a government body, often with the involvement of stakeholders, that describes a problem and broadly outlines how the problem will be addressed (Evans & Cvitanovic, [<reflink idref="bib10" id="ref15">10</reflink>]). In the public policy discourse, Bacchi ([<reflink idref="bib3" id="ref16">3</reflink>]) asserts that governments take policy decisions to "fix" certain "problems". A problematization approach to the development of policy helps in measuring the impact of the policy in terms of a theory of change (ToC). Such impacts in the context of AIED would be related to improved learning outcomes by reducing learning loss, student engagement, positive attitudes, improved capacity of teachers and educational administrators, and availability of technological infrastructure to access learning opportunities anywhere at anytime. Due to a problematization approach, a national AI in education policy must be contextual based on the local needs and resources (Pedro et al., [<reflink idref="bib19" id="ref17">19</reflink>]).</p> <p>Despite the importance of an AIED policy as highlighted in the aforementioned background, little information exists in the literature on how to write one. This might hinder the adoption of AIED. Therefore, this study aims to address this research gap by first discussing the features (see Section 2) and components (see Section 3) that constitute a good AIED policy. It then presents and details the steps for writing an AIED policy (see Section 4). As a result, this study could serve as a template for various stakeholders who are interested in writing an AIED policy in their context.</p> <hd id="AN0182583709-3">AI in Education Policy Features</hd> <p>Developing a good policy document relies on identifying the defining features of a well written one (Ivanič, [<reflink idref="bib14" id="ref18">14</reflink>]). A haphazard policy document that does not possess the essentials of what makes a policy document complete and inclusive will be deficient at the time of implementing it as it would fail to address and/or meet the requirements to make it effective. This is even more significant in the writing of a policy document for AIED due to the fact that the implications of having a policy that is lacking in one of the essential features may have dire consequences for learners, teachers, school administration personnel or any person or institution that delivers education. As education is a public good and the backbone to the multifaceted development of a country, an AIED policy needs to make sure that it satisfies the essentials of a good policy document. The following is a discussion of the features of a good AIED policy.</p> <hd id="AN0182583709-4">Contextual</hd> <p>A comprehensive AIED policy must be contextual and address the peculiar characteristics of the conditions of implementation so that it is relevant and needs-based. What works in an environment where AI has developed to a level where it has many users with the requisite skills and resources may not work in situations where AI is still in its infancy and users are not very well acquainted with the technology and may not even possess the devices required to run them at scale. This is particularly true in the African context, where it is observed that AI adoption in education is still in its fancy, calling for further attention in this regard (Adams, [<reflink idref="bib1" id="ref19">1</reflink>]).</p> <hd id="AN0182583709-5">Consultative</hd> <p>A good policy document has to be a result of a consultative process to capture the voices and aspirations of the various stakeholders who will make use of the policy once it is enacted. The eventual acceptance of policy demands and provisions will be maximized only if potential users and stakeholders are consulted about it right from its inception and their needs, views, and concerns are considered in the development of the policy document. The consultative process will also have to be undertaken in a very transparent manner in which the genuine interests of users and providers are reflected in good faith. It is in keeping with this cardinal requirement that AI transparency will reduce any potential bias, hence increasing its trustability and adoption among various stakeholders (Lawton, [<reflink idref="bib15" id="ref20">15</reflink>]). Transparency must be built within the AI policy and its implementation as AI deals with data storage, retrieval and use with potentially profound effects on individuals and/or groups (Nguyen et al., [<reflink idref="bib17" id="ref21">17</reflink>]; Tlili et al., [<reflink idref="bib23" id="ref22">23</reflink>]). For this reason, any AIED policy development and implementation must not be done in an obscure environment. The transparency should also be within the AIED policy assessment mechanisms to reveal the true effect of that policy.</p> <hd id="AN0182583709-6">Dynamic</hd> <p>A good policy document must be dynamic to allow revision, modifications and the incorporation of new developments in the field. This implies not only that policies are intrinsically processual, but as a whole they are not the simple sum of their components (actors, contexts, instruments, funds, etc.) (Capano, [<reflink idref="bib6" id="ref23">6</reflink>]). It is therefore important to keep in mind that an effect of a given policy should not be taken for granted, and policy makers should constantly try to enhance that effect through finding the adequate interconnections and patterns among policy components.</p> <p>This is particularly important in rapidly evolving fields such as AI. No policy document will be able to capture all the ramifications and applications of AI. It is therefore necessary to provide a framework that would allow the integration of new uses and requirements, so to say that "all policy is policy change" (Hogwood & Peters, [<reflink idref="bib13" id="ref24">13</reflink>]). One way of ensuring the timeliness of the policy is taking it through a periodic revision and updating process that would allow the incorporation of new developments, tools and procedures into a parent policy document. While updating a policy, the following questions should be considered: what has really changed? how do policies change? when do they change? why do they change? (Capano, [<reflink idref="bib6" id="ref25">6</reflink>]).</p> <hd id="AN0182583709-7">Implementable and Measurable</hd> <p>The developed AIED policy should be measurable in terms of its impact so that its significance can be assessed later on. It is therefore important that when developing an AI policy to be realistic and in line with the feature of the given context (as discussed in the contextual feature) where the AI will be developed instead of proposing something fancy and unreasonable, hence undoable in reality. Therefore, it is crucial to set achievable implementation targets for an AIED policy, as well as quantifiable performance indicators that can help to measure and evaluate the AIED policy objectives later on.</p> <hd id="AN0182583709-8">AI in Education Policy Components</hd> <p>A policy document, like any organic entity, should have clearly defined and identifiable parts to itself (Apthorpe, [<reflink idref="bib2" id="ref26">2</reflink>]). If any part is missing or not fully formed, it would affect its effectiveness negatively (Young & Quinn, [<reflink idref="bib28" id="ref27">28</reflink>]). A deficient policy document will affect the implementation of the provisions within it as what is not considered during its writing will be conspicuous by its absence during the implementation phase. Consideration of the components of a good policy document well ahead of its writing would ensure that the essential ingredients are considered and guide its development. AIED, as a rapidly unfolding area of technology as applied to education, will have to be guided by a policy document that clearly shows the essential parts that need to be considered. The following is a discussion of the components of a good AIED policy.</p> <hd id="AN0182583709-9">Needs Analysis of End Users and Considering Socio-Economic Variables</hd> <p>One of the components of a good AIED policy is that it sets out with a thorough analysis of the conditions of use (or nonuse) of AI within the education system that serves learners, teachers, school administration personnel, curriculum development staff, etc (Pham & Sampson, [<reflink idref="bib20" id="ref28">20</reflink>]). Doing so helps design relevant content and applications, and the exercise should begin with a needs analysis which will serve as a backdrop for the writing of a policy. In addition to the needs analysis of end users, the policy must also have sufficient background information on the socio-economic variables that will have an impact on the implementation of the policy and the perception of the end users toward it (Sen et al., [<reflink idref="bib22" id="ref29">22</reflink>]). An AIED policy development exercise that is not based on a sound analysis of needs and context is destined to either be irrelevant or even become a hindrance rather than a facilitator of learning. That is exactly why a wholesale adoption of an AIED policy document developed for one country is undesirable in another or even harmful. In addition, the very nature of AI and its rapid growth and unpredictability require that countries ground its adoption and use on the realities obtained in their countries rather than "import" any AI system and apply it in their schools and/or learning institutions.</p> <hd id="AN0182583709-10">Algorithm and Data Regulation</hd> <p>The second most important component of a good AIED policy is making sure that there is no bias, intended or unintended, in the algorithms and data used to train the AI system to be implemented. This is very important as, "... various studies have found rampant discrimination perpetuated by AI against certain groups of people" (Global Partnership of Sustainable Development Data, [<reflink idref="bib11" id="ref30">11</reflink>], p. 44). This can be avoided by selecting "... training data that is appropriately representative and large enough to counteract common types of machine learning bias, such as sample bias and prejudice bias" (Vashisht, [<reflink idref="bib25" id="ref31">25</reflink>]).</p> <p>Once algorithm and data bias are avoided, the next step is setting out the regulations for access and use of digital data of learners, teachers, school personnel, and school property to ensure data accuracy and security. This is very important as users of any AI system need reassurance that their data and identity do not fall into the wrong hands and result in damage to their personality and/or property. The policy has to spell out the tools and procedures through which data is accessed and used, with clear and predictable mechanisms built into the system to safeguard identity and data. Particularly, it is crucial to highlight what kind of data will be collected and the mechanisms of doing that. It should also specify how the data will be stored, who will have access to it, and how it is secured from being tampered with. The conditions of data transfer and disclosure to third parties (if this is necessary for legal and security reasons) must also be part of the implementation protocol. In this regard, one of the recommendations of UNESCO (Miao et al., [<reflink idref="bib16" id="ref32">16</reflink>], p. 32) is to "Establish data protection laws which make educational data collection and analysis visible, traceable, and auditable by teachers, students and parents: Formulate clear policies regarding data ownership, privacy and availability for the public good." As emphasized by the Centre for Intellectual Property and Information Technology Law (CIPIT, [<reflink idref="bib7" id="ref33">7</reflink>], p. 6), "While AI has great potential, it also presents significant ethical challenges for governments, developers and users. These include accountability, data bias, transparency, and socio-economic concerns such as social inequality."</p> <hd id="AN0182583709-11">Safe Adoption and Use</hd> <p>The third component of an AIED policy should deal with implementation guidelines which spell out how an adopted AI system is integrated safely with the school curriculum that intends to enrich. The policy should provide regulations that go beyond the typical ethical regulation (privacy, data regulation, etc.) to promote developing responsible AI-based educational systems. This means that the AI systems should augment humans (teachers, learners, school leaders, etc.) instead of replacing them, as well as how AI systems can ensure human dignity and have a positive impact, beyond education, on society. In this context, several policymakers (Berendt et al., [<reflink idref="bib5" id="ref34">5</reflink>]; Hagendorff, [<reflink idref="bib12" id="ref35">12</reflink>]; Nigam et al., [<reflink idref="bib18" id="ref36">18</reflink>]) highlighted the need to focus on the ethical guidelines of AI that align with societal values.</p> <p>Closely aligned with the integration of AI with curriculums is the training of teachers and other education personnel on the use of the AI system for maximum effectiveness and efficiency. A good number of teachers, especially in Africa and South East Asia, may not have sufficient level of qualifications and training to be able to use AI for teaching. School personnel too need data literacy training to handle a huge amount of personal data on students, teachers and school property at their disposal. The implementation guidelines should also be explicit on issues of access and inclusion in order to make sure that the AI system adopted is accessible to all learners despite differences in economic standing, geographic location, physical disabilities, gender, age, etc. (UNESCO, [<reflink idref="bib24" id="ref37">24</reflink>]).</p> <hd id="AN0182583709-12">Sustainability</hd> <p>When implemented, a comprehensive AIED policy should also consider other domains, such as economy, and culture, among others, to also ensure their sustainability (in addition to the sustainability of the education field) (European Parliament, [<reflink idref="bib8" id="ref38">8</reflink>]; European Union, [<reflink idref="bib9" id="ref39">9</reflink>]). For instance, when developing an AIED policy, it is important to consider energy consumption optimization, ecological footprint reduction, and employability inclusion and diversification.</p> <hd id="AN0182583709-13">Monitoring and Evaluation Scheme</hd> <p>A good AIED policy should also detail the monitoring and evaluation scheme of the AI system as part of the guidelines for implementation. UNESCO pointed out, "the development of monitoring and evaluation mechanisms to measure the impact of AI on education, teaching and learning, in order to provide a valid and robust evidence-based foundation for policy-making" (Miao et al., [<reflink idref="bib16" id="ref40">16</reflink>]). Particularly, an extra eye should be paid to the research methodology of evaluating the true effect of AIED coming from a particular policy, as several researchers reported that the existing methodologies are a black box and cannot tell much (Wang et al., [<reflink idref="bib26" id="ref41">26</reflink>]). The results of the monitoring and evaluation exercise will then feed into the revision of the policy at periodic intervals. This way the timeliness of the provisions of the policy shall be ensured.</p> <hd id="AN0182583709-14">AI in Education Policy Development Process</hd> <p>The process involved in developing an AIED policy is more or less similar to the process followed in the development of any other social policy (Baldock & Mitton, [<reflink idref="bib4" id="ref42">4</reflink>]). This is because the whole purpose of any policy is the regulation of socially significant interventions by a government or an institution for the betterment of the conditions of life and/or to govern a relationship between a client and a provider. In this connection, an AIED policy aspires to regulate the setting up and implementation of an AI system within the education sector of a country. The following are the suggested steps for writing an AIED policy.</p> <hd id="AN0182583709-15">Pre-Drafting Consultation</hd> <p>This is the initial activity that will bring together stakeholders to discuss how to go about developing the policy. As AIED is primarily concerned with education, the Ministry of Education (MoE) of a country can be the major stakeholder that initiates the consultation in collaboration with related ministries such as the Ministry of Science and Technology, Telecommunications, Planning, Finance, Labor and other government entities who may have a stake in the application of AI in schools and learning institutions. Teacher unions should also be consulted and join the discussion as they are one of the major stakeholders for AIED. The pre-drafting consultation should also include civil society organizations and advocacy groups that work in the areas of education, digital access, literacy, disability, women, etc. This is done to canvas opinions and input from diverse groups and enrich the policy that comes out of such consultation. The terms of reference for the writing team can also come from the pre-writing consultations. As mentioned above, the eventual buy-in of the policy will largely be determined by how consultative it has been during its development. The pre-drafting consultations should also focus on the type of policy to be developed in the context of the existing policies. For example, the AIED policy may take the form of: (i) a stand-alone policy, (ii) integrated with the national ICT in education policy, or (iii) take the form of a thematic policy by focusing on just one aspect of the AI in education policy.</p> <hd id="AN0182583709-16">Stakeholder Survey</hd> <p>Before drafting a policy, it is important to undertake a comprehensive baseline survey of the stakeholders (students, teachers, parents, school administrators, curriculum developers, public and private education providers). Such a survey must focus on a representative sample of the population and include questions related to socio-economic backgrounds, ICT infrastructure, access, and affordability, infrastructure at schools (including Internet bandwidth), attitudes of the stakeholders and analysis of skills and competencies related to understanding of AI. This survey will provide the evidence-base for drafting the policy and make it relevant and appropriate to the needs of stakeholders.</p> <hd id="AN0182583709-17">Writing Draft Policy</hd> <p>Actual writing will have to be done by a technical team entrusted with the task. The team can be composed of MoE, teacher unions, civil society, relevant ministries, and other actors known to be actively involved in digital and access issues. A technical group should not be too big as this will make it unwieldy to manage. However, the writing team should delineate the overall context within which the policy is to be situated and has to focus on legal and institutional frameworks. It must also make sure that the draft policy has the components discussed above to make sure that it is as comprehensive as possible. The draft policy must also include an implementation plan, covering specific tasks (specific objectives), timelines, responsible authority for implementation, sources of funding, targets and means of verification.</p> <hd id="AN0182583709-18">Discussion on Draft Policy</hd> <p>Just like in the pre-drafting phase, the discussion on the draft policy will have to be as broad-based as possible. The discussion may have to be conducted among many more participants than the pre-drafting consultation as the pre-drafting consultation would normally enlist people who are somehow knowledgeable about the subject or are assumed to be the ones who have a legal mandate to it. However, the consultation on the draft policy should allow the participation of a wider segment of potential "users" either directly or indirectly. For example, parents would be more interested in being part of the discussion on the draft document than in the pre-drafting consultation.</p> <p>The media will have a more visible role in this phase of the process as it can be used to encourage debate through interviews, panel discussions, print and electronic media feature articles, etc. The media can also be used to disseminate the policy among the population for wider discussion and buy-in. At the end of such an extensive discussion and dissemination exercise, the policy will be a document of compromise and the result of a lot of negotiations. This is so because no policy document can be a fully satisfactory piece of work for all stakeholders. At the same time, the compromise cannot be on the basic tenets of inclusion, transparency, and data security.</p> <hd id="AN0182583709-19">Adoption and Communication of Policy Document</hd> <p>Once discussions are conducted and comments and suggestions have enriched the draft policy document, the next step is going through the adoption of the policy in preparation for its implementation. This can be done by a legal entity entrusted with issuing approvals for policy instruments. Depending on the level of emphasis given to AIED policy by a government and the significance of the provisions contained in it for the welfare and security of the population, the approval can be given by a legislative body such as the parliament.</p> <p>At the time of adoption of the policy, the draft implementation plan must be thoroughly reviewed to allocate the budget needed for successful implementation of the policy. Many policies fail due to absence of budgetary provisions, and accountability of the implementing agencies. Communicating the policy decisions to the relevant implementation agencies and stakeholders is key to ensuring that the benefits of the policy reach everyone.</p> <hd id="AN0182583709-20">Policy Implementation Plan</hd> <p>The next phase in the development of an AIED policy is its implementation. This will have to be governed by an implementation plan that spells out how the policy will be rolled out. This segment of the policy development exercise will be done over an extended period of time starting from introducing the policy to putting the provisions in it into practice. This can be handled by an inter-sectoral monitoring body that will oversee what happens on the ground in schools and learning institutions. Actual implementation can also be done in phases as conditions on the ground may not be the same in all the schools in a country. Better endowed areas and schools may implement more of the provisions from the start, and less endowed ones may have to progress in stages. This may be necessary due to both financial and other reasons such as availability of internet connectivity, power, trained manpower, hardware, etc. Monitoring the implementation and communicating the outcomes to the stakeholders are important steps in the success of a policy. Monitoring helps in understanding the path toward progress and identifying where more interventions could be needed. An important aspect where continuous monitoring is required is regular capacity building of teachers to understand and integrate AI in teaching and learning. Stories of success must also be shared to promote deeper appreciation of policy interventions.</p> <hd id="AN0182583709-21">Policy Evaluation</hd> <p>Any policy will have to be monitored over a set period and appropriate adjustments need to be made depending on how it has served the purpose for which it was developed. A formal evaluation also needs to be made at the end of a period of implementation considered sufficient. This is necessary to see if the policy had achieved the purposes for which it was enacted and whether there would be a need for any amendment and/or change. It is recommended that the evaluation be done by a neutral body for the sake of objectivity, and the opinions of direct beneficiaries of the policy need to be considered. A reasonable time must be given to implementation agencies before evaluation may be initiated. A recommended timeline is to undertake evaluation after three years.</p> <p>To sum up, Figure 1 depicts the AIED policy development process that various stakeholders can refer to.</p> <p>Graph: Figure 1. AI in education policy development process.</p> <hd id="AN0182583709-22">Conclusions</hd> <p>The paradox of the need for adopting AI in educational systems to create new learning opportunities and at the same time the need for being well prepared to be protected from AI continues to emerge in education. This paradox can only be addressed by launching and implementing comprehensive AIED policies. While several policymakers urged to focus on AIED policy, scant information exists in the literature on how to write one. Therefore, this present study contributes to this research gap by discussing the steps (7 steps) of writing an AIED policy, as well as the features and components of a good policy. It revealed that a well-defined AIED policy is not a simple process, and should be thoroughly developed. Particularly, it should be context-driven instead of "blindly" driven, where multi-disciplinary stakeholders are involved. Additionally, a good AIED policy should think beyond AI as a technology only, to consider AI as an ecosystem. Moreover, a good AIED policy should also ensure the sustainability of other fields rather than mainly tackling education.</p> <p>This present study, therefore, can serve as a reference for various stakeholders who are interested in developing an AIED policy in their contexts. This can help to increase the safe adoption of AIED and keep up with the societal rapid transformation that is shaped by various technologies, including AI. Consequently, this can help to promote the achievement of United Nations' sustainable development goals.</p> <hd id="AN0182583709-23">Disclosure statement</hd> <p>No potential conflict of interest was reported by the author(s).</p> <hd id="AN0182583709-24">Data availability statement</hd> <p>No new data was created or analyzed in this study. Data sharing is not applicable to this article.</p> <ref id="AN0182583709-25"> <title> References </title> <blist> <bibl id="bib1" idref="ref1" type="bt">1</bibl> <bibtext> Adams, R. (2022). AI in Africa: Key concerns and policy considerations for the future of the continent. Africa Policy Research Institute (APRI). https://afripoli.org/ai-in-africa-key-concerns-and-policy-considerations-for-the-future-of-the-continent</bibtext> </blist> <blist> <bibl id="bib2" idref="ref2" type="bt">2</bibl> <bibtext> Apthorpe, R. (1997). Writing development policy and policy analysis plain or clear: on language, genre and power. In C. Shore & S. Wright (Eds.), Anthropology of Policy: Critical perspectives on Governance and Power (pp. 43 – 58). London : Routledge.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref3" type="bt">3</bibl> <bibtext> Bacchi, C. (2009). Analysing policy: What's the problem represented to be? Pearson Higher Education 5.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref4" type="bt">4</bibl> <bibtext> Baldock, J., & Mitton, L. (Eds.). (2011). Social policy. Oxford University Press.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref5" type="bt">5</bibl> <bibtext> Berendt, B., Littlejohn, A., & Blakemore, M. (2020). AI in education: Learner choice and fundamental rights. Learning, Media and Technology, 45 (3), 312 – 324. https://doi.org/10.1080/17439884.2020.1786399</bibtext> </blist> <blist> <bibl id="bib6" idref="ref6" type="bt">6</bibl> <bibtext> Capano, G. (2012). Policy dynamics and change: The never-ending puzzle. In E. Araral, S. Fritzen, M. Howlett, M. Ramesh, & X. Wu (Eds.), Routledge handbook of public policy (pp. 451 – 472). New York : Routledge.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref7" type="bt">7</bibl> <bibtext> Centre for Intellectual Property and Information Technology Law. (2023). State of artificial intelligence in Africa 2023 report. Kenya. https://cipit.strathmore.edu/state-of-artificial-intelligence-in-africa-2023-report/</bibtext> </blist> <blist> <bibl id="bib8" idref="ref38" type="bt">8</bibl> <bibtext> European Parliament. (2021). Report on artificial intelligence in education, culture and the audiovisual sector (2020/2017(INI)). Committee on Culture and Education. https://<ulink href="http://www.europarl.europa.eu/doceo/document/A-9-2021-0127%5fEN.html">www.europarl.europa.eu/doceo/document/A-9-2021-0127%5fEN.html</ulink></bibtext> </blist> <blist> <bibl id="bib9" idref="ref39" type="bt">9</bibl> <bibtext> European Union. (2019). Ethics guidelines for trustworthy AI. https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai</bibtext> </blist> <blist> <bibtext> Evans, M. C., & Cvitanovic, C. (2018). An introduction to achieving policy impact for early career researchers. Palgrave Communications, 4 (1), 1 – 12. https://doi.org/10.1057/s41599-018-0144-2</bibtext> </blist> <blist> <bibtext> Global Partnership of Sustainable Development Data. (2023). Artificial intelligence practitioners' Guide: Kenya. https://www.bmz-digital.global/wp-content/uploads/2023/04/GIZ-2023-Artificial-Intelligence-Practicioners-Guide-Kenya.pdf</bibtext> </blist> <blist> <bibtext> Hagendorff, T. (2020). The ethics of AI ethics: An evaluation of guidelines. Minds and Machines, 30 (1), 99 – 120. https://doi.org/10.1007/s11023-020-09517-8</bibtext> </blist> <blist> <bibtext> Hogwood, B. W., & Peters, G. B. (1983). Policy dynamics. St. Martin Press.</bibtext> </blist> <blist> <bibtext> Ivanič, R. (2004). Discourses of writing and learning to write. Language and Education, 18 (3), 220 – 245. https://doi.org/10.1080/09500780408666877</bibtext> </blist> <blist> <bibtext> Lawton, G. (2023). AI transparency: What is it and why do we need it? https://<ulink href="http://www.techtarget.com/searchcio/tip/AI-transparency-What-is-it-and-why-do-we-need-it#:~:text=%22At%20the%20end%20of%20the,in%20the%20field%20of%20AI">www.techtarget.com/searchcio/tip/AI-transparency-What-is-it-and-why-do-we-need-it#:~:text=%22At%20the%20end%20of%20the,in%20the%20field%20of%20AI</ulink></bibtext> </blist> <blist> <bibtext> Miao, F., Holmes, W., Huang, R., & Zhang, H. (2021). AI and education: Guidance for policy-makers. United Nations Educational, Scientific and Cultural Organization. https://unesdoc.unesco.org/ark:/48223/pf0000376709</bibtext> </blist> <blist> <bibtext> Nguyen, A., Ngo, H. N., Hong, Y., Dang, B., & Nguyen, B. P. T. (2023). Ethical principles for artificial intelligence in education. Education and Information Technologies, 28 (4), 4221 – 4241. https://doi.org/10.1007/s10639-022-11316-w</bibtext> </blist> <blist> <bibtext> Nigam, A., Pasricha, R., Singh, T., & Churi, P. (2021). A systematic review on ai-based proctoring systems: Past, present and future. Education and Information Technologies, 26 (5), 6421 – 6445. https://doi.org/10.1007/s10639-021-10597-x</bibtext> </blist> <blist> <bibtext> Pedro, F., Subosa, M., Rivas, A., & Valverde, P. (2019). Artificial intelligence in education: Challenges and opportunities for sustainable development. UNESCO. <ulink href="http://repositorio.minedu.gob.pe/bitstream/handle/20.500.12799/6533/Artificial%20intelligence%20in%20education%20challenges%20and%20opportunities%20for%20sustainable%20development.pdf">http://repositorio.minedu.gob.pe/bitstream/handle/20.500.12799/6533/Artificial%20intelligence%20in%20education%20challenges%20and%20opportunities%20for%20sustainable%20development.pdf</ulink></bibtext> </blist> <blist> <bibtext> Pham, S. T., & Sampson, P. M. (2022). The development of artificial intelligence in education: A review in context. Journal of Computer Assisted Learning, 38 (5), 1408 – 1421. https://doi.org/10.1111/jcal.12687</bibtext> </blist> <blist> <bibtext> Schiff, D. (2022). Education for AI, not AI for education: The role of education and ethics in national AI policy strategies. International Journal of Artificial Intelligence in Education, 32 (3), 527 – 563. https://doi.org/10.1007/s40593-021-00270-2</bibtext> </blist> <blist> <bibtext> Sen, D., Günay, M. E., & Tunç, K. M. (2019). Forecasting annual natural gas consumption using socio-economic indicators for making future policies. Energy, 173, 1106 – 1118. https://doi.org/10.1016/j.energy.2019.02.130</bibtext> </blist> <blist> <bibtext> Tlili, A., Huang, R., Mustafa, M. Y., Zhao, J., Bozkurt, A., Xu, L., Wang, H., Salha, S., Altinay, F., Affouneh, S., Burgos, D. (2023). Speaking of transparency: Are all Artificial Intelligence (AI) literature reviews in education transparent? Journal of Applied Learning & Teaching, 6 (2), 44 – 55. https://doi.org/10.37074/jalt.2023.6.2.15</bibtext> </blist> <blist> <bibtext> United Nations Educational, Scientific and Cultural Organization. (2019). Beijing consensus on artificial intelligence and education [ Paper presentation ]. Outcome document of the International Conference on Artificial Intelligence and Education, Planning Education in the AI Era: Lead the Leap, Beijing. https://unesdoc.unesco.org/ark:/48223/pf0000368303</bibtext> </blist> <blist> <bibtext> Vashisht, R. (2021). How to reduce machine learning bias. https://medium.com/atoti/how-to-reduce-machine-learning-bias-eb24923dd18e</bibtext> </blist> <blist> <bibtext> Wang, H., Tlili, A., Huang, R., Cai, Z., Li, M., Cheng, Z., Yang, D., Li, M., Zhu, X., & Fei, C. (2023). Examining the applications of intelligent tutoring systems in real educational contexts: A systematic literature review from the social experiment perspective. Education and Information Technologies, 1 – 36. https://doi.org/10.1007/s10639-022-11555-x</bibtext> </blist> <blist> <bibtext> Yan, L., Sha, L., Zhao, L., Li, Y., Martinez-Maldonado, R., Chen, G., Li, X., Jin, Y., & Gašević, D. (2023). Practical and ethical challenges of large language models in education: A systematic literature review. British Journal of Educational Technology, 55 (1), 90 – 112. https://doi.org/10.1111/bjet.13370</bibtext> </blist> <blist> <bibtext> Young, E., & Quinn, L. (2002). Writing effective public policy papers. Open Society Institute.</bibtext> </blist> </ref> <aug> <p>By Awol Endris; Ahmed Tlili; Ronghuai Huang; Lin Xu; TingWen Chang and Sanjaya Mishra</p> <p>Reported by Author; Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib24" firstref="ref8"></nolink> <nolink nlid="nl2" bibid="bib21" firstref="ref9"></nolink> <nolink nlid="nl3" bibid="bib27" firstref="ref10"></nolink> <nolink nlid="nl4" bibid="bib16" firstref="ref11"></nolink> <nolink nlid="nl5" bibid="bib10" firstref="ref15"></nolink> <nolink nlid="nl6" bibid="bib19" firstref="ref17"></nolink> <nolink nlid="nl7" bibid="bib14" firstref="ref18"></nolink> <nolink nlid="nl8" bibid="bib15" firstref="ref20"></nolink> <nolink nlid="nl9" bibid="bib17" firstref="ref21"></nolink> <nolink nlid="nl10" bibid="bib23" firstref="ref22"></nolink> <nolink nlid="nl11" bibid="bib13" firstref="ref24"></nolink> <nolink nlid="nl12" bibid="bib28" firstref="ref27"></nolink> <nolink nlid="nl13" bibid="bib20" firstref="ref28"></nolink> <nolink nlid="nl14" bibid="bib22" firstref="ref29"></nolink> <nolink nlid="nl15" bibid="bib11" firstref="ref30"></nolink> <nolink nlid="nl16" bibid="bib25" firstref="ref31"></nolink> <nolink nlid="nl17" bibid="bib12" firstref="ref35"></nolink> <nolink nlid="nl18" bibid="bib18" firstref="ref36"></nolink> <nolink nlid="nl19" bibid="bib26" firstref="ref41"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1459780
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Features, Components and Processes of Developing Policy for Artificial Intelligence in Education (AIED): Toward a Sustainable AIED Development and Adoption
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Awol+Endris%22">Awol Endris</searchLink><br /><searchLink fieldCode="AR" term="%22Ahmed+Tlili%22">Ahmed Tlili</searchLink><br /><searchLink fieldCode="AR" term="%22Ronghuai+Huang%22">Ronghuai Huang</searchLink><br /><searchLink fieldCode="AR" term="%22Lin+Xu%22">Lin Xu</searchLink><br /><searchLink fieldCode="AR" term="%22TingWen+Chang%22">TingWen Chang</searchLink><br /><searchLink fieldCode="AR" term="%22Sanjaya+Mishra%22">Sanjaya Mishra</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Leadership+and+Policy+in+Schools%22"><i>Leadership and Policy in Schools</i></searchLink>. 2025 24(1):233-241.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 9
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2025
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Descriptive
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Policy+Formation%22">Policy Formation</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+Intelligence%22">Artificial Intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Technology+Uses+in+Education%22">Technology Uses in Education</searchLink><br /><searchLink fieldCode="DE" term="%22Sustainable+Development%22">Sustainable Development</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+Policy%22">Educational Policy</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+Improvement%22">Educational Improvement</searchLink><br /><searchLink fieldCode="DE" term="%22Ethics%22">Ethics</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/15700763.2024.2312999
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1570-0763<br />1744-5043
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Governments and private sectors are now putting in place the needed resources and infrastructure to harness the power of emerging technologies in education. One of these technologies is Artificial Intelligence (AI) which gained increasing attention due to its potential to enhance learning and teaching experiences, hence achieving better learning outcomes. However, AIED also comes with several concerns that raise continuous questions about its safe and effective adoption. The application of AIED needs to be planned and executed properly. One of the basic requirements for this to happen is that a comprehensive national policy on the use of AIED is put in place to guide its implementation and evaluate its effectiveness. Limited information exists in the literature on how to write an AIED policy. To address this research gap, this study therefore discusses the features, major components and processes that countries are advised to adopt for a comprehensive AIED policy development. Specifically, this study highlights four features for a good AIED policy, namely contextual, consultative, dynamic, and, implementable and measurable. It further proposes seven steps for an AIED policy development, namely (1) pre-drafting consultation, (2) stakeholder survey, (3) writing draft policy, (4) discussion on draft policy, (5) adoption and communication of policy document, (6) policy implementation plan, and (7) policy evaluation.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2025
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1459780
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1459780
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/15700763.2024.2312999
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 9
        StartPage: 233
    Subjects:
      – SubjectFull: Policy Formation
        Type: general
      – SubjectFull: Artificial Intelligence
        Type: general
      – SubjectFull: Technology Uses in Education
        Type: general
      – SubjectFull: Sustainable Development
        Type: general
      – SubjectFull: Educational Policy
        Type: general
      – SubjectFull: Educational Improvement
        Type: general
      – SubjectFull: Ethics
        Type: general
    Titles:
      – TitleFull: Features, Components and Processes of Developing Policy for Artificial Intelligence in Education (AIED): Toward a Sustainable AIED Development and Adoption
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Awol Endris
      – PersonEntity:
          Name:
            NameFull: Ahmed Tlili
      – PersonEntity:
          Name:
            NameFull: Ronghuai Huang
      – PersonEntity:
          Name:
            NameFull: Lin Xu
      – PersonEntity:
          Name:
            NameFull: TingWen Chang
      – PersonEntity:
          Name:
            NameFull: Sanjaya Mishra
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 1570-0763
            – Type: issn-electronic
              Value: 1744-5043
          Numbering:
            – Type: volume
              Value: 24
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: Leadership and Policy in Schools
              Type: main
ResultId 1