AI & Education

ANNOTATED BIBLIOGRAPHY

Click on the triangle bullet point for the full annotation!

Alasadi, Eman A., and Charles R. Baiz. “Generative AI in Education and Research: Opportunities, Concerns, and Solutions.” Journal of Chemical Education,  vol. 100, no. 8, 2023, pp. 2965-2971. https://doi.org/10.1021/acs.jchemed.3c00323

The paper goes over the potential role of AI in transforming the future education system and the importance of readers to understand the full benefit of incorporating AI as supporting tools for education. Alasadi and Baiz go on to mention how instructors could use AI to tailor their course materials to fit the needs of individual students that contain students’ target learning objectives (as STEM and Humanities Majors), different learning styles, and language barriers. There is also the prediction of a paradigm shift of educators needing training on how to use and incorporate AI effectively into their teaching as the world rapidly shifts to adopt AI usage in employment and future professions. The authors of this paper seem to have very pro-AI integration into education when they mention that they would be discussing how to implement AI into education rather than discussing whether or not the education system should be implementing AI. The authors further drive this point in by using ChatGPT themselves to generate responses to the 6 main concerns listed in the paper. While it could mean that the authors are very forward and future looking, the authors fail to mention any potential lay offs of educators that may come with the efficacy of integrating AI. The paper mentions using AI to support ESL students and researchers due to shortage of bilingual teachers, but does not mention the humanistic implications of using AI to overcome the language barrier rather than transitioning into hiring more bilingual educators. In one of the responses to the 6 main concerns, the AI response includes how hiring multiple educators in a classroom is cost prohibitive in the financial perspective, and offers using AI tools as a financially friendly option. This pro-AI approach paper is highly relevant to our project that concerns the humanistic impacts and implications that AI integration may have on students and educators as we see a rapid shift in education due to institutions adapting AI as part of their learning tools. 

Babu, Sankaran, et al. “AI in the Hands of Gen Z: ChatGPT in Education and Employment.” Social Sciences & Humanities Open, vol. 10, 2024, article 101163, https://doi.org/10.1016/j.ssaho.2024.101163

The authors argue that Gen Z university students’ ChatGPT usage rates are significantly influenced by a combination of generational identity, students’ attitudes/views, and adoption tendencies. They test these relationships using survey data from university students (n = 307) analyzed with a structural equation model (SEM) in Amos. This source is important because it gives you a clean behavioral/adoption model for explaining why students choose to use ChatGPT (not just whether they use it). For my thesis that student adoption and dependency are driven by attitudes and perceived usefulness (and can create uneven patterns of use), this study provides a tested framework tying student views and adoption tendency to actual usage rate. It also helps set up our public vs. private comparison question because the authors note their sample focuses on public-university students and point to the need to examine private-university contexts next.

Baek, Clare, Tamara Tate, and Mark Warschauer. “‘ChatGPT seems too good to be true’: College students’ use and perceptions of generative AI.” Computers and Education: Artificial Intelligence, vol. 7, 2024, article 100294, https://doi.org/10.1016/j.caeai.2024.100294

The authors argue that U.S. college students’ ChatGPT use and attitudes are shaped by social and institutional factors (like demographics and school policy) and that AI can both support and disadvantage different groups in higher education. They use a large student sample (N = 1001) with regression analyses plus thematic analysis/NLP on open-ended responses to connect usage patterns and sentiments to characteristics like major, institution type, and institutional policy. This source is important because it directly links who uses ChatGPT, for what tasks, and why equity and policy issues are exactly the kind of “student perception + institutional context” connection your project needs. For my thesis that student reliance on AI depends heavily on policy clarity and student background, this article provides measurable evidence that institution type/policy and demographics predict different kinds of ChatGPT use (general, writing, programming). It also supports our risk/policy questions by showing students’ fear of punishment and concerns about displacement and fairness in access.

Bauer, Elisabeth, et al. “Looking beyond the hype: Understanding the effects of AI on learning.” Educational Psychology Review, vol. 37, no. 2, 2025, https://doi.org/10.1007/s10648-025-10020-8

The authors argue that AI’s impact on learning is not automatically positive. It depends on how AI is implemented, and poorly implemented AI can even create “inversion effects” like over reliance and reduced cognitive engagement. Instead of collecting new survey data, the article synthesizes existing empirical and theoretical research trends, and introduces the ISAR model (inversion, substitution, augmentation, redefinition) to classify different AI effects. This source is important because it gives our project a credible framework for interpreting student perceptions especially around dependence, performance expectations, and risks without treating “AI use” as inherently good or bad. For my thesis that student reliance on AI can increase efficiency but also risks weakening learning when students outsource thinking, this article provides the concept of over reliance/inversion and explains why that risk emerges. It also supports our argument that universities should respond with clear policies and AI literacy support for students and educators because these are moderating conditions for  whether AI substitutes for instruction or actually improves learning outcomes.

Chan, Cecilia Ka Yuk, and Louisa H. Y. Tsi. “Will Generative AI Replace Teachers in Higher Education? A Study of Teacher and Student Perceptions.” Studies in Educational Evaluation, vol. 83, 2024, article 101395, https://doi.org/10.1016/j.stueduc.2024.101395

This article argues that students generally see generative AI as useful for learning but also have concerns about academic integrity and ethical risks. The study uses survey responses and qualitative feedback from students in higher education to examine perceptions, benefits and challenges of AI. This resource is important because it provides direct evidence of how students understand and react to AI in academic settings. For the research, this article helps explain student attitudes toward AI, perceived risks, and how policy awareness may influence student behavior and use of AI tools.

Deng, Ruiqi, et al. “Does CHATGPT Enhance Student Learning? A Systematic Review and Meta-Analysis of Experimental Studies.” ScienceDirect, 12 Dec. 2024, www.sciencedirect.com/science/article/pii/S0360131524002380

The authors argue that ChatGPT has statistically significant positive effects on student learning across experimental studies, with its impact varying depending on subject area and education level. They conducted a systematic review of experimental studies that used controlled educational interventions that compared ChatGPT-supported learning to traditional education. This source is important because it gives us high-level, combined evidence, creating a reliable foundation for evaluating AI’s academic impact. This study provides an overall baseline of how ChatGPT is influencing student learning and performance. By comparing it to Indonesian student data, we can determine whether similar patterns appear in Indonesian universities, which will strengthen our analysis.

Fauziddin, Mohammad, et al. “The Impact of AI on the Future of Education in Indonesia.” Educative Jurnal Ilmiah Pendidikan, vol. 3, no. 1, 16 Jan. 2025, pp. 1–16, https://doi.org/10.70437/educative.v3i1.828

This article, completed by researchers from different Indonesian universities, is a literature review of articles that discuss AI and Education in Indonesia specifically, between 2014 and 2024. After reviewing the papers, the authors synthesized the findings into subthemes. Some findings especially relevant was that 60% of Indonesian students face difficulties due to lack of educational resources of high quality. Another was that only 30% of Indonesian schools have the internet capabilities necessary to support any technology-based or online-based learning. Despite these and concerns about algorithmic bias, these researchers seem optimistic about the potential of AI in helping improve student outcomes. As our dataset is composed of Indonesian respondents, this paper is incredibly helpful in providing us with necessary context regarding the current challenges and opportunities for technology education in Indonesia and allows us to frame our understanding of the responses better.

Giannakos, Michail, et al. “The Promise and Challenges of Generative AI in Education”, 2 Sept. 2024, www.tandfonline.com/doi/full/10.1080/0144929X.2024.2394886

This article addresses the positive and negative use cases of GenAI technologies, specifically Large Language Model technology, in the context of learning technologies and education. The authors interviewed experts who work in learning technology research across five countries, who have shared their views and reflections on GenAI technologies, and they reference their insights and commentaries throughout the article. This resource is important because as GenAI technologies, especially Large Language Models, grow increasingly used and relevant in education for both students and teachers, it is important to raise awareness of AI advantages and potential disadvantages. In our dataset, we have information about the types of Large Language Models that Indonesian students use for their learning and education, and this article provides a focus on LLM models. Moreover, our dataset contains a section where students rate if they’re likely to continue using AI in the future or recommend them to their friends, and this article also looks towards using AI in the future which can help provide insightful commentary for our research.

Kasneci, Enkelejda, et al. “ChatGPT for Good? On Opportunities and Challenges of Large Language Models for Education.” Learning and Individual Differences, vol. 103, 2023, article 102274. https://doi.org/10.1016/j.lindif.2023.102274

This article argues that large language models like ChatGPT can enhance learning efficiency and accessibility while also introducing risks such as bias and over-reliance. The authors support this argument by reviewing existing research on AI applications in education and analyzing both opportunities and challenges. This resource is important because it provides a balanced perspective on the benefits and risks of AI in education. This article helps evaluate how AI influences learning, student perceptions and concerns about the long-term impact of AI on academic development.

Klimova, Blanka, and Marcel Pikhart. “Exploring the Effects of Artificial Intelligence on Student and Academic Well-Being in Higher Education: A Mini-Review.” Frontiers in Psychology, U.S. National Library of Medicine, 3 Feb. 2025, pmc.ncbi.nlm.nih.gov/articles/PMC11830699/

This article assesses how AI can affect student well-being, focusing on mental health, social interactions, and learning experiences, and highlights for a well-balanced AI integration to support academic success and student well-being. To support their arguments, the article synthesizes 24 supporting literature pieces. This resource is important because it highlights potential long-lasting consequences of AI usage such as diminished emotional intelligence and technostress. Since our dataset has measured how much students use AI, this article can help us discuss the potential and dangerous effects of AI usage. This article can also support our discussion of AI benefits to educational contexts, such as the association between the use of AI and higher happiness scores.

Knoth, Stefan, and Daniela Tolzin. “AI Literacy and Prompt Engineering: An Interdisciplinary Project in Higher Education.” Computers and Education: Artificial Intelligence, vol. 6, 2024, article 100225. https://doi.org/10.1016/j.caeai.2024.100225

The authors argue that effective prompt engineering is an important skill for goal directed use of LLM tools in education and that students’ AI literacy helps determine how well they can prompt and adapt these tools. They use a skill-based approach that connects prompt quality to output quality and adds qualitative insights about students’ intuitive behaviors when interacting with LLM systems. This source is important because it supports the idea that “AI use” isn’t one thing students get very different results depending on their AI literacy and prompting skill, which can affect satisfaction and perceived performance. For my thesis that student expectations and perceived effectiveness of AI tools depend on their ability to use them well, this article gives a direct mechanism like better prompts to better outputs, and AI literacy helps produce better prompts. It supports our questions about technical difficulties/training needs by arguing for integrating AI education content into curricula so students can use tools like ChatGPT more effectively and responsibly.

Mytra, Prima, et al. “Digital Literacy and digital ethics of university students in the era of Artificial Intelligence: A study at Universitas Islam Ahmad Dahlan Sinjai.” JTMT: Journal Tadris Matematika, vol. 6, no. 2, 25 Nov. 2025, pp. 20–28, https://doi.org/10.47435/jtmt.v6i2.4116

The authors argue that an institution’s level of digital literacy and ethics significantly influences how effectively and responsibly its students use AI tools. They support this argument by conducting in-depth interviews and observations with students from three faculties at Universitas Islam Ahmad Dahlan Sinjai in Indonesia. This source is important because it provides university-specific information about their critical and ethical level of digital literacy and their use of AI in the academic setting. This will allow us to interpret our data not only by AI access, but also by students’ ability to use AI effectively and responsibly.

Pireci Sejdiu, Nora, and Sejdi Sejdiu. “The Quiet Transformation of Higher Education in the AI Era.” Open Research Europe, U.S. National Library of Medicine, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC12438950/

The authors go over the ethical concerns and uncertainty in potential implications of AI adaptation in educational institutions. There are ethical dilemmas and concerns that educators are currently faced with when assessing student learning. By late 2023, more than half of the US student population reported using AI for their assignments and 86% of the users reporting that their AI use was undetected by instructors. While new technological innovations such as calculators have touched one domain, AI influences a wider range of education and learning, everything from writing, solving problems, and having conversations. This paper also mentions AI-driven personalization as great learning tools for inclusive education as the other paper, “Generative AI in Education and Research: Opportunities, Concerns, and Solutions.” by Alasadi and Baiz mentioned. While mentioning some positives of deeper and more tailored student learning creating an environment for more inclusive education, there are risks of marginalized communities being left further behind in the widespread adaptation of AI use in education. There is a profound concern in AI integration further driving the inequalities in education by widening the gaps between students who have full access to AI and students who may not have reliable internet access or do not feel comfortable with technology. There is also a need for flexibility in institutional policies regarding what is considered academic integrity with the large paradigm shift. Sejdiu et al. take a comprehensive and humanistic standpoint on how AI integration may cause dynamic shifts in the educational system and calls for a holistic re-evaluation of the education system, while not stating specifically how the policy changes should be made. 

Raharjo, Riris Setyaningrum, and Samsul Huda Rohmadi. “Artificial Intelligence in Indonesian education: A critical review of ethical considerations, Implementation Challenges, and Educational Management Perspectives.” At-Tarbawi: Jurnal Kajian Kependidikan Islam, vol. 10, no. 1, 5 July 2025, pp. 50–68, https://doi.org/10.22515/attarbawi.v10i1.12141

The authors critically review the challenges of implementing AI in education in Indonesia and find solutions for ethical concerns and institutional management challenges. They support their argument with a literature review that investigates ethical frameworks and policy discussions of AI implementation in the Indonesian education system. This source is important because it is specific to the integration of AI in the Indonesian education system. This study provides background on ethical and cultural factors that influence the integration of AI and how it affects student performance in Indonesian universities. 

Sari, Desy K., et al. “Measuring Artificial Intelligence Literacy: The Perspective of Indonesian Higher Education Students.” Journal of Pedagogical Research, vol. 9, no. 2, 21 Mar. 2025, pp. 143-157, https://doi.org/10.33902/jpr.202531879

This study also discusses the Indonesian education system, but in a slightly different way. Researchers from several different universities in Indonesia surveyed 542 university students that were in their last year and asked them several survey questions, including AI literary questions. The study then shows the difference in these scores by gender of the participants, by region, by device owned, and other demographic features. This study is very useful in helping us pre-plan which patterns to look for in our own dataset. For example, one of the main findings in this study is that the majority of the participants have very low or low levels of AI literacy. For us, this is important to know as such perspectives can clearly bias questions regarding perception and implementation of AI in educational systems, as there is a foundational lack of understanding. This helps us bring in a more nuanced perspective when seeing our own results and allows us to tell the narrative better than simply relying on our current knowledge and only one dataset. 

Vieriu, Aniella Mihaela, and Gabriel Petrea. “The Impact of Artificial Intelligence (AI) on Students’ Academic Development.” Education Sciences, vol. 15, no. 3, 11 Mar. 2025, p. 343, https://doi.org/10.3390/educsci15030343

This study that was done by the National University of Science and Technology POLITEHNICA Bucharest asked 85 second-year students in aerospace and medical engineering programs 11 questions regarding their use of AI and their perceptions surrounding it. We believe this an interesting sample to study as it provides another Non-American exposure to attitudes surrounding AI, and includes a university student sample in STEM fields, which we can compare to the STEM students in our dataset. An overwhelming majority of the respondents in this study (88.2%) reported using virtual assistants, and 95.6% reported using AI for educational purposes. The thematic insights were positive, with most students believing that AI use contributes to their academic success while only 2% suggest it could lead to a decline in performance. However, some common concerns that were brought up was the lack of trust in AI generated content. This study serves as a useful resource in understanding key risks of integrating AI in educational systems and serves as a benchmark that allows us to compare and contrast our findings against. 

Younas, Muhammad, et al. “The Impact of Artificial Intelligence-Based Learning Tools in Academic Innovation: A Review of Deep Seek, GPT, and Gemini (2020–2025).” Frontiers, 16 Nov. 2025, www.frontiersin.org/journals/education/articles/10.3389/feduc.2025.1689205/full

This article underlines the significant improvements yet challenges that AI learning tools can offer to learning experiences. The article cites multiple peer-reviewed research papers that were published between 2020 and 2025, offering up-to-date sources of AI perspectives. This resource is important for our team to reinforce the advantages and disadvantages of AI usage in education by using multiple supporting sources. Some learning tools this article focuses on are ChatGPT, Deep Seek, and Gemini, which is relevant to our dataset as we have records of the types of tools that Indonesian students have used. This resource will also assist with our team’s project by supporting our claims about the critical issues, challenges, and benefits of AI, in both the teaching and learning.

Zawacki-Richter, Olaf, et al. “Systematic Review of Research on Artificial Intelligence Applications in Higher Education: Where Are the Educators?” International Journal of Educational Technology in Higher Education, vol. 16, no. 39, 2019. https://doi.org/10.1186/s41239-019-0171-0

This article argues that artificial intelligence has significant potential in higher education but should complement rather than replace human teaching. The study uses a systematic review of research on AI applications in education to examine trends, benefits and challenges. This resource is important because it provides a broad overview of how AI has been used in higher education and the evolving role of educators. This article helps place the study within a broader context and supports the analysis of how AI may influence learning, teaching, and student perceptions over time.

Other Citations