ChatGPT and large language models in academia: opportunities and challenges
Open accessAbstract
The introduction of large language models (LLMs) that allow iterative “chat” in late 2022 is a paradigm shift that enables generation of text often indistinguishable from that written by humans. LLM-based chatbots have immense potential to improve academic work efficiency, but the ethical implications of their fair use and inherent bias must be considered. In this editorial, we discuss this technology from the academic’s perspective with regard to its limitations and utility for academic writing, education, and programming. We end with our stance with regard to using LLMs and chatbots in academia, which is summarized as (1) we must find ways to effectively use them, (2) their use does not constitute plagiarism (although they may produce plagiarized text), (3) we must quantify their bias, (4) users must be cautious of their poor accuracy, and (5) the future is bright for their application to research and as an academic tool.
AI snapshot
Written from the paper’s full text.
TL;DR
This editorial argues that LLM-based chatbots such as ChatGPT should be embraced in academia for writing, education, and programming, while cautioning about bias, hallucination, and poor accuracy. The authors conclude that LLM use does not constitute plagiarism, that bias metrics are needed, and that users remain responsible for validating outputs.
Key findings
The authors’ stance is that (1) LLMs must be embraced, (2) their use does not constitute plagiarism although they may produce plagiarized text, (3) their bias must be quantified, (4) users must be cautious of poor accuracy, and (5) the future is bright for their application to research and as an academic tool.
- ChatGPT can present false information as truth: when asked to help find relevant publications to cite, it “proceeded to make up a list of five entirely fictional publications”; on repeat trials it sometimes listed one or two real papers, but other times fabricated a paper or returned an irrelevant one.
- On scalability and accessibility, the paper reports ChatGPT gained 100 million active users in 2 months, with an estimated 13 million daily queries, compared with Google search’s estimated 8.5 billion daily queries; ChatGPT offers a free tier and a $20/month premium plan. It also notes BLOOM as an open access 176B parameter multilingual LLM, ChatGPT as a 175 billion parameter LLM, speculation that OpenAI had access to 10,000 Nvidia V100s costing close to $5 million, and an estimate that training BERT (described as a 6 billion parameter LLM) would produce 21 to 78 metric tons of CO2.
- For academic writing, LLMs may particularly help non-native English speakers; the paper cites that 49% of postdoctoral fellows trained in the United States were born overseas, and in engineering, math, and computer science, 60% of PhDs were awarded to international students. For programming, ChatGPT can help write, interpret, debug, simplify, and translate code, but it may present incorrect code as correct, miss edge cases, and is more reliable for smaller blocks than larger complex programs.
Methods
Design: editorial/commentary, not an empirical study or systematic review.
- Data or sample: no primary data; the paper draws on published literature, illustrative examples, and the authors’ experiences.
- Main analysis: narrative discussion of opportunities and challenges across academic writing, education, programming, bias, and scalability/accessibility.
Limitations
The paper does not report a formal limitations section for its own approach. Evident from the text: it is an editorial/commentary, so its examples are illustrative and include personal experiences rather than systematic evidence.
- The paper states LLM limitations: hallucination/false information, bias from training data, poor accuracy, misrepresentation of capabilities, and inability to anticipate edge cases in code.
- The paper notes unresolved ethical/legal uncertainty around LLM use in academic writing and grants, and that detection of cheating remains an unsolved challenge for educators.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 18 cited sources checked · 0 supported · 0 flagged
- Findingreasoning
ChatGPT can present false information as truth (hallucination).
“A fundamental challenge with LLM-based chatbots is that they can present false information as truth, also known as “hallucination”.”
- Also stated in: § Background · plain
“For example, when asked if it could help find relevant publications to cite in a review paper, ChatGPT confirmed that it could, but then proceeded to make up a list of five entirely fictional publications.”
Quote in the paperSubstantial claim - Also stated in: § Background · plain
- Findingreasoning
ChatGPT and Bard can misrepresent their capabilities.
“In fact, ChatGPT and Bard can misrepresent their capabilities.”
Quote in the paperSubstantial claim - Premisecitation
LLMs may present false information as truth.
“For example, LLMs may adopt bias, perpetuate stereotypes in the training dataset, and present false information as truth.”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [3] Sallam M. ChatGPT utility in healthcare education, research, and practice: systematic review on the promising perspectives and valid concerns. Healthcare. 2023;11(6):887. https://doi.org/10.3390/healthcare11060887. source
Not checked in the free run.
- cites [4] Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepaño C, et al. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLOS Digit Health. 2023;2:e0000198. source
Not checked in the free run.
- cites [5] Liebrenz M, Schleifer R, Buadze A, Bhugra D, Smith A. Generating scholarly content with ChatGPT: ethical challenges for medical publishing. Lancet Digit Health. 2023;5:e105–6. source
Not checked in the free run.
- Interpretationreasoning
LLM chatbots have limited utility in domains such as the medical field where facts are essential.
“These issues limit the utility of LLM chatbots in some domains, such as the medical field, where facts are essential to ensure the best health outcomes.”
Quote in the paperNot a substantial claim - Interpretationreasoning
An answer generated by a large language model can be formatted correctly but not necessarily be factual.
“Thus, an answer generated by a large language model can be formatted correctly but not necessarily be factual.”
Quote in the paperSubstantial claim
- Premisecitation
Over reliance on ChatGPT may hinder students' ability to develop critical skills, such as writing.
“Over reliance may hinder students ability to develop critical skills, such as writing [21].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [21] Shidiq M. The use of artificial intelligence-based chat-gpt and its challenges for the world of education; from the viewpoint of the development of creative writing skills. Proc Int Conf Educ Soc Humanity. 2023;1(1):353–7. ISSN 2986-5832.
Not checked in the free run.
- Interpretationreasoning
Using LLMs as an editing tool with pre-written text is unlikely to raise ethical concerns such as plagiarism.
“Consequently, employing LMMs with pre-written text is unlikely to raise ethical concerns such as plagiarism, which may arise when using prompt-based text generation by LMMs.”
Quote in the paperSubstantial claim - Premisecitation
AI/ML solutions deployed in healthcare can exacerbate and perpetuate systematic biases and disparities for marginalized groups.
“Unethically and irresponsibly designed AI/ML solutions deployed in the healthcare setting can exacerbate and perpetuate systematic biases and disparities for those from marginalized groups [27–29].”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [27] Vyas DA, Eisenstein LG, Jones DS. Hidden in Plain Sight - Reconsidering the Use of Race Correction in Clinical Algorithms. N Engl J Med. 2020;383:874–82. source
Not checked in the free run.
- cites [28] Gijsberts CM, Groenewegen KA, Hoefer IE, Eijkemans MJC, Asselbergs FW, Anderson TJ, et al. Race/Ethnic Differences in the Associations of the Framingham Risk Factors with Carotid IMT and Cardiovascular Events. PLoS ONE. 2015;10:e0132321. source
Not checked in the free run.
- cites [29] Obermeyer Z, Powers B, Vogeli C, Mullainathan S. Dissecting Racial Bias in an Algorithm used to Manage the Health of Populations. Science. 2019;366:447–53. source
Not checked in the free run.
- Premisecitation
Training BERT (a 6 billion parameter LLM) would produce between 21 to 78 metric tons of CO2 to train to completion.
“A recent study estimated that the training of BERT (a 6 billion parameter LLM) would produce between 21 to 78 metric tons of CO2 to train to completion [14].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [14] Dodge J, Prewitt T, Des Combes RT, Odmark E, Schwartz R, Strubell E, Luccioni AS, Smith NA, DeCario N, Buchanan W. Measuring the Carbon Intensity of AI in Cloud Instances. arXiv. 2022;2206.05229. source
Not checked in the free run.
- Premisecitation
Until root-causes of encoded biases of LLMs are addressed, LLMs for clinical applications will suffer the same fate, propagating biases.
“Until the root-causes of the encoded biases of LLMs are addressed, LLMs for clinical applications will suffer the same fate, propagating biases [30, 31].”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [30] Neeley T, Ruper S. Timnit Gebru: 'SILENCED No More' on AI Bias and The Harms of Large Language Models. Harvard Business School Case. 2022;422–085.
Not checked in the free run.
- cites [31] Bender EM, Gebru T, McMillan-Major A, Shmitchell S. On the dangers of stochastic parrots: can language models be too big? In: Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency. New York, NY, USA: ACM; 2021. p. 610–23. source
Not checked in the free run.
- Classified as probably not a substantial claim. 23
- May combine two claims. 11
Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Findingown dataIn one author's experience, ChatGPT can help provide ideas for interesting research directions and suggest logical reasons for research in grant writing.
“In one author's experience experimenting with how ChatGPT might help write grants, ChatGPT can help provide some ideas for interesting directions, even suggesting sound logical reasons that we need to do research that could find their way into the significance section of NIH grants.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
FindingreasoningIn the authors' experience, none of the text produced by ChatGPT made it into the final grant proposal.
“However, grants must convey many complex ideas in a short amount of space, and as such, since much of the text from ChatGPT is so general and often repetitive, in our experience none of the text produced by ChatGPT made it into the final grant proposal.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Findingown dataChatGPT can provide general text that might be logically used in parts of a grant, such as the aims page.
“ChatGPT can even provide general text that might be logically used in parts of the grant, such as the aims page.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Findingnone givenChatGPT typically responds in a confident, seemingly knowing manner, whether the provided information is accurate or not.
“A more important concern regarding students learning from a LLM is that it typically responds in a confident, and seemingly knowing manner, whether the provided information is accurate or not.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Findingown dataChatGPT can suggest specific experiments that might test a hypothesis of interest.
“It can suggest specific experiments that might test a hypothesis of interest.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Findingown dataChatGPT helped one author overcome writer's block.
“A surprising effect for one author of this editorial was that it helped overcome writer's block in that simply having some text on the page was sufficient to move forward with editing.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Findingnone givenThe model can generate correctly formatted information such as references that do not exist.
“The model can generate correctly formatted information such as references, that do not exist.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based). May combine two claims.
Findingnone givenChatGPT will misrepresent its knowledge; instead of responding "I don't know", it will readily provide fabricated information with apparent confidence.
“ChatGPT will misrepresent its knowledge; instead of responding “I don’t know”, it will readily provide fabricated information with apparent confidence.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationLLMs may adopt bias and perpetuate stereotypes present in their training data.
“For example, LLMs may adopt bias, perpetuate stereotypes in the training dataset, and present false information as truth.”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [3] Sallam M. ChatGPT utility in healthcare education, research, and practice: systematic review on the promising perspectives and valid concerns. Healthcare. 2023;11(6):887. https://doi.org/10.3390/healthcare11060887. source
Not checked in the free run.
- cites [4] Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepaño C, et al. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLOS Digit Health. 2023;2:e0000198. source
Not checked in the free run.
- cites [5] Liebrenz M, Schleifer R, Buadze A, Bhugra D, Smith A. Generating scholarly content with ChatGPT: ethical challenges for medical publishing. Lancet Digit Health. 2023;5:e105–6. source
Not checked in the free run.
Classified as probably not a substantial claim.
InterpretationreasoningBasing treatment plans on outputs from ChatGPT, at least in its current state, would lead to a large proportion of misdiagnoses and ineffective treatments that could cause patient harm.
“For example, it would not be wise to base treatment plans on outputs from ChatGPT, at least in its current state, because this would lead to a large proportion of misdiagnoses and ineffective treatments that could cause patient harm.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
InterpretationreasoningDeploying AI models at scale is a challenge that future AI technologies will have to face.
“Deploying AI models at scale is a challenge that future AI technologies will have to face even for tech giants such as Google (Bard power by LaMDA) and Microsoft (AI powered Bing) that have announced AI integration within their respective search engines.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Interpretationnone givenChatGPT may unpredictably present incorrect code as being correct.
“In line with warnings for general use, ChatGPT may somewhat unpredictably present incorrect code as being correct.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationStudents might easily use ChatGPT dishonestly, either to cheat in homework or exams, or to write reports/essays without citation in a manner that could be construed as plagiarism.
“Concerns have arisen regarding how students might easily use ChatGPT dishonestly, either to cheat in the completion of homework or exams, or to write reports/essays without citation in a manner that could be construed as plagiarism [22].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [22] Neumann M, Rauschenberger M, Schön E-M. “We Need To Talk About ChatGPT”: The Future of AI and Higher Education. 2023:4. https://doi.org/10.25968/opus-2467. source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.
InterpretationreasoningChatGPT is more likely to be successful in accurately writing smaller blocks of code, whereas its reliability in writing larger/more complex programs is questionable.
“Currently, ChatGPT is more likely to be successful in accurately writing smaller blocks of code, whereas its reliability in writing larger/more complex programs (e.g. a software package) is questionable.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).
InterpretationreasoningCode written by ChatGPT may offer a useful starting point.
“Ultimately, code written by ChatGPT may offer a useful starting point, however both the code and comments that it generates should be checked and validated by both the programmer and users to confirm that it fully satisfies the intended purpose.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.
Interpretationnone givenChatGPT can facilitate computer programming in learning to code, writing new code, interpreting existing code, debugging existing code, increasing code efficiency, and translating code between languages.
“ChatGPT also promises to facilitate computer programming in terms of (1) learning to code or use specific packages, libraries, or frameworks, (2) writing new code, (3) interpreting existing code, (4) debugging existing code, (5) increasing the compactness or efficiency of code, or (6) translating code from one programming language to another.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationLLMs provide benefits and opportunities in education for students by assisting with research and academic writing or using them as an interactive study guide.
“LLMs provide benefits and opportunities in education for students by assisting with research and academic writing [15] or using them as an interactive study guide which may include generating practice exams and being provided immediate feedback [16, 17].”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [15] Kasneci E, Sessler K, Küchemann S, Bannert M, Dementieva D, Fischer F, et al. ChatGPT for good? On opportunities and challenges of large language models for education. Learn Individ Differ. 2023;103:102274. source
Not checked in the free run.
- cites [16] MacNeil S, Tran A, Mogil D, Bernstein S, Ross E, Huang Z. Generating Diverse Code Explanations using the GPT-3 Large Language Model. In: Vahrenhold J, Fisler K, Hauswirth M, Franklin D, editors. Proceedings of the 2022 ACM Conference on International Computing Education Research - Volume 2. New York, NY, USA: ACM; 2022. p. 37–9. source
Not checked in the free run.
- cites [17] Tate TP, Doroudi S, Ritchie D, Xu Y, Mark Warschauer UCI. Educational Research and AI-Generated Writing: Confronting the Coming Tsunami. 2023.
Not checked in the free run.
Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).
InterpretationreasoningPaywalled access to AI could further increase the divide between the wealthy and the poor.
“While AI has the potential to revolutionize education and academic research, paywalled access could further increase the divide between the wealthy and the poor.”
Quote in the paperNot a substantial claim- Rests on an assumption: AI has the potential to revolutionize education and academic research.
May combine two claims.
PremisecitationData produced from ChatGPT may be inaccurate or biased.
“Data produced from ChatGPT may be inaccurate or biased [23], which users need to be aware of to ensure these deficiencies are not propagated in their works.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [23] Baidoo-Anu D, Owusu Ansah L. Education in the Era of Generative Artificial Intelligence (AI): Understanding the Potential Benefits of ChatGPT in Promoting Teaching and Learning. 2023. Available at SSRN: https://ssrn.com/abstract=4337484 or https://doi.org/10.2139/ssrn.4337484. source
Not checked in the free run.
Classified as probably not a substantial claim. May combine two claims.
InterpretationreasoningLLMs increase the efficiency of academic writing, education, and programming.
“This editorial provides several examples and perspectives on how LLMs increase the efficiency of academic writing, education, and programming.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S4 (based).
Interpretationnone givenIn many contexts, LLMs have low accuracy.
“In many contexts, LLMs have low accuracy.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
InterpretationreasoningOutput from ChatGPT will match existing text on the internet, particularly if the sample is small enough.
“Output from ChatGPT will match existing text on the internet, particularly if the sample is small enough.”
Quote in the paperNot a substantial claim- Rests on an assumption: The output of ChatGPT is from most publicly available text on the internet until 2021.
Probably not a substantial claim: may not pass S4 (based).
Premisenone givenChatGPT does not do anything without a prompt.
“However, ChatGPT does not do anything without a prompt.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).
InterpretationreasoningWhen LLMs for clinical applications are socially, consciously, and transparently designed with bias considerations, they can become an additional tool for promoting equity and improving access to care.
“When LLMs for clinical applications are socially, consciously, and transparently designed with such considerations, they can become an additional tool for promoting equity and improving access to care.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim. May combine two claims.
InterpretationreasoningThe performance of ChatGPT and other LLMs mimics the available text, and they are as biased as their training data.
“The performance of ChatGPT and other LLMs therefore mimics the available text, and they are as biased as their training data.”
Quote in the paperSubstantial claim- Rests on an assumption: The output of ChatGPT is from most publicly available text on the internet until 2021.
- Rests on an assumption: ChatGPT is known to perpetuate stereotypes such as nurses being female and doctors being male, and this bias comes from the training data.
Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).
InterpretationreasoningIn a strict sense, LLMs like ChatGPT do produce plagiarism.
“Thus, in a strict sense, LLMs like ChatGPT do produce plagiarism.”
Quote in the paperNot a substantial claim- Rests on an assumption: The output of ChatGPT is from most publicly available text on the internet until 2021.
Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.
Interpretationnone givenAs companies race to innovate with LLMs, encoded biases will be amplified, and harm will manifest.
“As companies race to innovate with LLMs like ChatGPT and others, encoded biases will be amplified, and harm will manifest.”
Quote in the paperNot a substantial claim
The paper relies on this without support.
Effective academic writing is essential to the success and quality of academic publications and is a major challenge facing international postdocs and students.
“Effective academic writing plays an essential role in the success and quality of academic publications, nevertheless, it is one of the major challenges facing international postdocs and students.”
The paper relies on this without support.
The current grant review system is subjective; some proposals scored very poorly in one study section while nearly the same proposal at another study section is funded.
“Despite this massive investment, the current review system is subjective; for example, some proposals sent to one study section are scored very poorly, while nearly the same proposal at another study section is funded.”
The paper relies on this without support.
ChatGPT has already been applied to help write grant proposals.
“ChatGPT has undoubtedly already been applied to help write grant proposals.”
The paper relies on this without support.
ChatGPT is a 175 billion parameter LLM.
“As a reminder, ChatGPT is a 175 billion parameter LLM and this only considers the cost of training and not any cost related to generating text in response to potentially billions of daily queries.”
The paper relies on this without support.
The output of ChatGPT is from most publicly available text on the internet until 2021.
“The output of ChatGPT is from most publicly available text on the internet until 2021.”
The paper relies on this without support.
ChatGPT is known to perpetuate stereotypes such as nurses being female and doctors being male, and this bias comes from the training data.
“For example, ChatGPT is known to perpetuate stereotypes such as nurses being female and doctors being male, and this bias comes from the training data.”
The paper relies on this without support.
The use of ChatGPT and other LLMs by students is inevitable.
“As the use of ChatGPT and other LLMs by students is inevitable, educators will need to change the pedagogy and assessments in education, such as integrating the use of ChatGPT in lessons [26], and a focus more on the process rather than the final product for assessment.”
The paper relies on this without support.
Detection of cheaters remains an unsolved challenge for educators.
“While the detection of cheaters remains an unsolved challenge for educators, if we start from the assumption that students are allowed, if not encouraged, to utilize ChatGPT, strategies could be adopted to account for this.”
The paper relies on this without support.
It is difficult to define metrics that assess the level of bias in training data and model outputs.
“It is difficult to define metrics that assess the level of bias in the training data and in the model outputs.”
The paper relies on this without support.
AI has the potential to revolutionize education and academic research.
“While AI has the potential to revolutionize education and academic research, paywalled access could further increase the divide between the wealthy and the poor.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
GenAI can automate routine administrative tasks in higher education, such as consolidation of grades, creating performance summaries of students, crafting evaluation reports, and maintaining student enrolment and attendance records.
“using GenAI to automate routine administrative tasks in higher education, such as consolidation of grades, creating performance summaries of students, crafting out evaluation reports, and maintaining student enrolment and attendance records (Meyer et al., 2023).”
From Higher Education's Generative Artificial Intelligence Paradox: The Meaning of Chatbot Mania (2024) · cites this paper as Meyer et al., 2023
The Education section (page not supplied) says ChatGPT can facilitate educators’ assessment and evaluation, which partly supports the reference to evaluation tasks. It does not describe automating grade consolidation, student performance summaries, evaluation reports, or enrolment and attendance records; searches for administrative tasks and student records found no such passage. (LLM, conf=0.82)
Experimental — accuracy not yet measuredLLMs can act as a thought partner for students and enhance understanding by providing detailed explanations and step-by-step solutions.
“Third, LLMs can act as a thought partner for students and enhance understanding by providing detailed explanations and step-by-step solutions (Mollick and Mollick 2022, Kasneci et al. 2023, Meyer et al. 2023, Extance 2023).”
From AI Meets the Classroom: When Do Large Language Models Harm Learning? (2024) · cites this paper as Meyer et al. 2023
The Education section describes LLMs as interactive study guides that give students answers and immediate feedback. The Assistance with programming section says ChatGPT can act as an interactive instructor, explaining code line by line and offering solutions to coding questions. These passages support the thought-partner idea, but the paper’s detailed explanations and solutions are described specifically for programming, not as a general student-learning capability. (LLM, conf=0.82)
Experimental — accuracy not yet measuredThere has been a notable upswing in discussions among scholars and academicians regarding the transformative potential of generative AI, particularly ChatGPT, in higher education and the risks associated with it.
“there has been a notable upswing in discussions among scholars and academicians regarding the transformative potential of generative AI, particularly ChatGPT, in higher education and the risks associated with it (Dalalah & Dalalah, 2023; Meyer et al., 2023; Peters et al., 2023; Yilmaz & Yilmaz, 2023a).”
From Is it harmful or helpful? Examining the causes and consequences of generative AI usage among university students (2024) · cites this paper as Meyer et al., 2023
A self-referential loop in LLMs can lead to the propagation of biases, increased homogeneity in generated data, and ultimately hindered performance.
“The consequence of a self-referential loop in LLMs can lead to several problematic outcomes, including the propagation of biases [30], increased homogeneity in generated data, and ultimately, hindered performance.”
From Large Language Models and User Trust: Consequence of Self-Referential Learning Loop and the Deskilling of Health Care Professionals (2024) · cites this paper as [30]
ChatGPT is considered the fastest-growing technology in history, amassing an estimated 120 million monthly active users and 13 million daily queries in the three months following its public launch.
“ChatGPT is considered the fastest-growing technology in history: in just three months following its public launch, it amassed an estimated 120 million monthly active users [16] with an estimated 13 million daily queries [49], surpassing all other applications [64].”
From A systematic literature review of empirical research on ChatGPT in education (2024) · cites this paper as [49]
A fundamental challenge with LLM-based chatbots, including ChatGPT, is the accuracy and quality of the provided information and responses, as they provide false information as truth, a phenomenon often referred to as hallucination.
“a fundamental challenge with LLM-based chatbots, including ChatGPT, is the accuracy and quality of the provided information and responses, as they provide false information as truth—a phenomenon often referred to as “hallucination” [3, 49]”
From A systematic literature review of empirical research on ChatGPT in education (2024) · cites this paper as [49]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Reference [9] is identified as a preprint but cited as a journal publication
MinorThe text contrasts preprint [9] with final publication [4], but reference [9] is cited as a PLOS Digital Health article and has the same DOI as reference [4]. The cited entries therefore do not document separate preprint and final versions.
“In one case, ChatGPT was listed as an author in the preprint [9] but missing as an author in the final publication [4].”
“9. Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, et al. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLOS Digital Health. 2023;2(2):e0000198. 10.1371/journal.pdig.0000198. PMC9931230 36812645”
Cites
ChatGPT for good? On opportunities and challenges of large language models for education
Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., et al. (2023) · Learning and Individual Differences · doi:10.1016/j.lindif.2023.102274 · Cited by 7,326 · cited 1× · see citing works
“LLMs provide benefits and opportunities in education for students by assisting with research and academic writing [15] or using them as an interactive study guide which may include generating practice exams and being provided immediate feedback [16, 17].”
BackgroundDissecting racial bias in an algorithm used to manage the health of populations
Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019) · Science · doi:10.1126/science.aax2342 · Cited by 7,137 · cited 1× · see citing works
“Unethically and irresponsibly designed AI/ML solutions deployed in the healthcare setting can exacerbate and perpetuate systematic biases and disparities for those from marginalized groups [27–29].”
BackgroundOn the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988 · see citing works
Kung, T. H., Cheatham, M., Medenilla, A., Sillos, C., De Leon, L., Elepaño, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., Maningo, J., et al. (2023) · PLOS Digital Health · doi:10.1371/journal.pdig.0000198 · Cited by 4,115 · cited 1× · see citing works
“In one case, ChatGPT was listed as an author in the preprint [9] but missing as an author in the final publication [4].”
BackgroundSallam, M. (2023) · Healthcare · doi:10.3390/healthcare11060887 · Cited by 3,126 · cited 1× · see citing works
“On the other hand, there are also unsettling ethical issues to be considered [3–5].”
BackgroundChatting and cheating: Ensuring academic integrity in the era of ChatGPT
Cotton, D. R. E., Cotton, P. A., & Shipway, J. R. (2024) · Innovations in Education and Teaching International · doi:10.1080/14703297.2023.2190148 · Cited by 3,087 · see citing works
Chatting and Cheating. Ensuring academic integrity in the era of ChatGPT
Cotton, D., Cotton, P., & Shipway, J. R. (2023) · doi:10.35542/osf.io/mrz8h · Cited by 3,079 · see citing works
Baidoo-Anu, D., & Owusu Ansah, L. (2023) · doi:10.2139/ssrn.4337484 · Cited by 3,004 · cited 1× · see citing works
“Data produced from ChatGPT may be inaccurate or biased [23], which users need to be aware of to ensure these deficiencies are not propagated in their works.”
BackgroundGilson, A., Safranek, C. W., Huang, T., Socrates, V., Chi, L., Taylor, R. A., & Chartash, D. (2023) · JMIR Medical Education · doi:10.2196/45312 · Cited by 2,078 · cited 1× · see citing works
“In higher education, LLM-based chatbots may be used to increase student engagement, facilitate group activities, create interactive learning tools, and provide immediate feedback and assessment [18–20].”
BAİDOO-ANU, D., & OWUSU ANSAH, L. (2023) · Journal of AI · doi:10.61969/jai.1337500 · Cited by 1,826 · see citing works
Hidden in Plain Sight — Reconsidering the Use of Race Correction in Clinical Algorithms
Vyas, D. A., Eisenstein, L. G., & Jones, D. S. (2020) · New England Journal of Medicine · doi:10.1056/nejmms2004740 · Cited by 1,731 · see citing works
ChatGPT is fun, but not an author
Thorp, H. H. (2023) · Science · doi:10.1126/science.adg7879 · Cited by 1,437 · cited 1× · see citing works
“Science has declared that no AI generated text or figures are allowed in their published papers, and ChatGPT cannot be named as an author in their journal [10].”
BackgroundChatGPT listed as author on research papers: many scientists disapprove
Stokel-Walker, C. (2023) · Nature · doi:10.1038/d41586-023-00107-z · Cited by 868 · cited 1× · see citing works
“Indeed, we have already seen manuscripts where ChatGPT has been listed as author [6–8].”
BackgroundTools such as ChatGPT threaten transparent science; here are our ground rules for their use
(2023) · Nature · doi:10.1038/d41586-023-00191-1 · Cited by 605 · see citing works
A Conversation on Artificial Intelligence, Chatbots, and Plagiarism in Higher Education
King, M. R., & chatGPT (2023) · Cellular and Molecular Bioengineering · doi:10.1007/s12195-022-00754-8 · Cited by 602 · cited 1× · see citing works
“In a third example, a researcher presents the results of a conversation with ChatGPT around AI, chatbots using LLMs, and plagiarism in higher education [8].”
BackgroundGenerating scholarly content with ChatGPT: Ethical challenges for medical publishing
Liebrenz, M., Schleifer, R., Buadze, A., Bhugra, D., & Smith, A. (2023) · The Lancet Digital Health · doi:10.1016/s2589-7500(23)00019-5 · Cited by 594 · cited 1× · see citing works
“On the other hand, there are also unsettling ethical issues to be considered [3–5].”
BackgroundMeasuring the Carbon Intensity of AI in Cloud Instances
Dodge, J., Prewitt, T., Tachet des Combes, R., Odmark, E., Schwartz, R., Strubell, E., Luccioni, A. S., Smith, N. A., DeCario, N., & Buchanan, W. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533234 · Cited by 257 · cited 1× · see citing works
“A recent study estimated that the training of BERT (a 6 billion parameter LLM) would produce between 21 to 78 metric tons of CO 2 to train to completion [14].”
Background“We Need To Talk About ChatGPT”: The Future of AI and Higher Education
Neumann, M., Rauschenberger, M., & Schön, E.-M. (2023) · Hochschule Hannover · doi:10.25968/opus-2467 · Cited by 212 · see citing works
Rapamycin in the context of Pascal's Wager: generative pre-trained transformer perspective
Generative Pre-trained Transformer, C., & Zhavoronkov, A. (2022) · Oncoscience · doi:10.18632/oncoscience.571 · Cited by 203 · see citing works
Generating Diverse Code Explanations using the GPT-3 Large Language Model
MacNeil, S., Tran, A., Mogil, D., Bernstein, S., Ross, E., & Huang, Z. (2022) · Proceedings of the 2022 ACM Conference on International Computing Education Research - Volume 2 · doi:10.1145/3501709.3544280 · Cited by 200 · cited 1× · see citing works
“LLMs provide benefits and opportunities in education for students by assisting with research and academic writing [15] or using them as an interactive study guide which may include generating practice exams and being provided immediate feedback [16, 17].”
BackgroundGijsberts, C. M., Groenewegen, K. A., Hoefer, I. E., Eijkemans, M. J. C., Asselbergs, F. W., Anderson, T. J., Britton, A. R., Dekker, J. M., Engström, G., Evans, G. W., et al. (2015) · PLOS ONE · doi:10.1371/journal.pone.0132321 · Cited by 197 · see citing works
Human Language Understanding & Reasoning
Manning, C. D. (2022) · Daedalus · doi:10.1162/daed_a_01905 · Cited by 173 · cited 1× · see citing works
“ChatGPT is powered by a LLM, a type of deep learning model that emerged around 2018 [2].”
MethodEducational Research and AI-Generated Writing: Confronting the Coming Tsunami
Tate, T. P., Doroudi, S., Ritchie, D., Xu, Y., & uci, M. W. (2023) · doi:10.35542/osf.io/4mec3 · Cited by 137 · see citing works
Assessing the quality of student-generated short answer questions using GPT-3
Moore, S., Nguyen, H. A., Bier, N., Domadia, T., & Stamper, J. (2022) · Lecture Notes in Computer Science · doi:10.1007/978-3-031-16290-9_18 · Cited by 93 · see citing works
Algorithmic Fairness and AI Justice in Addressing Health Equity
Park, Y., Singh, M., Koski, E., Sow, D., Scheufele, E. L., & Bright, T. J. (2022) · Health informatics · doi:10.1007/978-3-031-07912-2_15 · Cited by 4 · see citing works
Self-citation
Flagged first, then by citations
Research works printed but not linked: 4
- Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, et al. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLOS Digital Health. 2023;2(2):e0000198. https://doi.org/10.1371/journal.pdig.0000198. · DOI · Paper page
- Shidiq M. The use of artificial intelligence-based chat-gpt and its challenges for the world of education; from the viewpoint of the development of creative writing skills. Proc Int Conf Educ Soc Humanity. 2023;1(1):353–7. ISSN 2986-5832.
- Trust T, Whalen J, Mouza C. Editorial: ChatGPT: Challenges, opportunities, and implications for teacher education. Contemp Issues Technol Teacher Educ. 2023;23(1):1–23.
- Dijkstra R, Genç Z, Kayal S, Kamps J. Reading Comprehension Quiz Generation using Generative Pre-trained Transformers. 2022.
Cited by
Sorted by citations
Using artificial intelligence in academic writing and research: An essential productivity tool
Khalifa, M., & Albadawy, M. (2024) · Computer Methods and Programs in Biomedicine Update · doi:10.1016/j.cmpbup.2024.100145 · Cited by 580 · see citing works
Impact of AI assistance on student agency
Darvishi, A., Khosravi, H., Sadiq, S., Gašević, D., & Siemens, G. (2024) · Computers & Education · doi:10.1016/j.compedu.2023.104967 · Cited by 548 · see citing works
“…educators in various aspects of pedagogy, enhancing productivity, generating detailed feedback, and reshaping the peer review process while also raising noteworthy considerations related to ethics, biases, and transparency (Cooper, 2023; Donker, 2023; Li, Patel, & Du, 2023; Meyer et al., 2023).”
BackgroundAbbas, M., Jam, F. A., & Khan, T. I. (2024) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-024-00444-7 · Cited by 531 · see citing works
“…and its potentially harmful or beneficial consequences is critical for educators, policymakers, and students, as it can help the development of effective strategies to integrate generative AI technologies into the learning process and control their misuse in higher education (Meyer et al., 2023).”
BackgroundA multinational study on the factors influencing university students’ attitudes and usage of ChatGPT
Abdaljaleel, M., Barakat, M. M., Alsanafi, M., Salim, N. A., Abazid, H., Malaeb, D. N., Mohammed, A. H., Hassan, B. A. R., Wayyes, A. M., Farhan, S. S., et al. (2024) · Scientific Reports · doi:10.1038/s41598-024-52549-8 · Cited by 354 · see citing works
“Several previous studies indicated the potential utility of ChatGPT as a prime example among other LLMs in higher education 1,18,26,63–65 .”
BackgroundMustafa, M. Y., Tlili, A., Lampropoulos, G., Huang, R., Jandrić, P., Zhao, J., Salha, S., Xu, L., Panda, S., Kinshuk, et al. (2024) · Smart Learning Environments · doi:10.1186/s40561-024-00350-5 · Cited by 235 · see citing works
“Recent studies have highlighted the opportunities that these techniques can bring in teaching and learning but also noted the challenges and the barriers that need to be addressed (Kas-neci et al., 2023; Meyer et al., 2023).”
A systematic literature review of empirical research on ChatGPT in education
Albadarin, Y., Saqr, M., Pope, N., & Tukiainen, M. (2024) · Discover Education · doi:10.1007/s44217-024-00138-2 · Cited by 191 · see citing works
“ChatGPT is considered the fastest-growing technology in history: in just three months following its public launch, it amassed an estimated 120 million monthly active users [16] with an estimated 13 million daily queries [49], surpassing all other applications [64].”
InfluentialKestin, G., Miller, K., Klales, A., Milbourne, T., & Ponti, G. (2025) · Scientific Reports · doi:10.1038/s41598-025-97652-6 · Cited by 214 · see citing works
Choudhury, A., & Chaudhry, Z. (2024) · Journal of Medical Internet Research · doi:10.2196/56764 · Cited by 159 · see citing works
“The consequence of a self-referential loop in LLMs can lead to several problematic outcomes, including the propagation of biases [30], increased homogeneity in generated data, and ultimately, hindered performance.”
A Design Space for Intelligent and Interactive Writing Assistants
Lee, M., Gero, K. I., Chung, J. J. Y., Shum, S. B., Raheja, V., Shen, H., Venugopalan, S., Wambsganss, T., Zhou, D., Alghamdi, E. A., et al. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642697 · Cited by 136 · see citing works
Higher Education's Generative Artificial Intelligence Paradox: The Meaning of Chatbot Mania
Rudolph, J., Mohamed Ismail, F. M., & Popenici, S. (2024) · Journal of University Teaching and Learning Practice · doi:10.53761/54fs5e77 · Cited by 128 · see citing works
“…processes (Javaid et al., 2023), using GenAI to automate routine administrative tasks in higher education, such as consolidation of grades, creating performance summaries of students, crafting out evaluation reports, and maintaining student enrolment and attendance records (Meyer et al., 2023).”
MethodBest Practices for Using AI Tools as an Author, Peer Reviewer, or Editor
Leung, T. I., de Azevedo Cardoso, T., Mavragani, A., & Eysenbach, G. (2023) · Journal of Medical Internet Research · doi:10.2196/51584 · Cited by 127 · see citing works
“At JMIR Publications, we adhere to this guidance of transparency and disclosure; we do not endorse a ban on generative AI in peer review, which can be counterproductive in various ways [14,35].”
Jo, H. (2023) · Telematics and Informatics · doi:10.1016/j.tele.2023.102067 · Cited by 127 · see citing works
Evangelista, E. D. L. (2025) · Contemporary Educational Technology · doi:10.30935/cedtech/15775 · Cited by 124 · see citing works
“Educators are growingly concerned that students will employ ChatGPT to create written assignments (Halaweh, 2023; Meyer et al., 2023).”
Large Language Model‐Based Chatbots in Higher Education
Yigci, D., Eryilmaz, M., Yetisen, A. K., Tasoglu, S., & Ozcan, A. (2025) · Advanced Intelligent Systems · doi:10.1002/aisy.202400429 · Cited by 113 · see citing works
Miao, J., Thongprayoon, C., Suppadungsuk, S., Garcia Valencia, O. A., Qureshi, F., & Cheungpasitporn, W. (2023) · Clinics and Practice · doi:10.3390/clinpract14010008 · Cited by 111 · see citing works
“Rather than debating whether AI should be used, the primary focus should revolve around how it can be harnessed responsibly and effectively [74].”
BackgroundPutman, T., Schaper, K., Matentzoglu, N., Rubinetti, V., Alquaddoomi, F. S., Cox, C., Caufield, J., Elsarboukh, G., Gehrke, S., Hegde, H. B., et al. (2023) · Nucleic Acids Research · doi:10.1093/nar/gkad1082 · Cited by 116 · see citing works
Bias in Machine Learning: A Literature Review
Mavrogiorgos, K., Kiourtis, A., Mavrogiorgou, A., Menychtas, A., & Kyriazis, D. (2024) · Applied Sciences · doi:10.3390/app14198860 · Cited by 109 · see citing works
An analysis of large language models: their impact and potential applications
Bharathi Mohan, G., Prasanna Kumar, R., Vishal Krishh, P., Keerthinathan, A., Lavanya, G., Meghana, M. K. U., Sulthana, S., & Doss, S. (2024) · Knowledge and Information Systems · doi:10.1007/s10115-024-02120-8 · Cited by 99 · see citing works
Generative Pre-Trained Transformer (GPT) in Research: A Systematic Review on Data Augmentation
Sufi, F. K. (2024) · Information · doi:10.3390/info15020099 · Cited by 97 · see citing works
Large Language Models in Computer Science Education: A Systematic Literature Review
Raihan, M. N., Siddiq, M. L., Santos, J. C. S., & Zampieri, M. (2025) · ACM Technical Symposium on Computer Science Education (SIGCSE) · doi:10.1145/3641554.3701863 · Cited by 125 · see citing works
“However, current methods, including plagiarism detection software, often fail to deliver reliable performance when handling output from a recently introduced LLM [7].”
Chen, K., Tallant, A. C., & Selig, I. (2025) · Information and Learning Sciences · doi:10.1108/ils-10-2023-0160 · Cited by 104 · see citing works
Exploring the role of large language models in the scientific method: from hypothesis to discovery
Zhang, Y., Khan, S. A., Mahmud, A., Yang, H., Lavin, A., Levin, M. G., Frey, J. G., Dunnmon, J., Evans, J., Bundy, A., et al. (2025) · npj Artificial Intelligence · doi:10.1038/s44387-025-00019-5 · Cited by 109 · see citing works
Large Language Model Prompting Techniques for Advancement in Clinical Medicine
Shah, K., Xu, A. Y., Sharma, Y. K., Daher, M., McDonald, C. L., Diebo, B. G., & Daniels, A. H. (2024) · Journal of Clinical Medicine · doi:10.3390/jcm13175101 · Cited by 94 · see citing works
“…can take several measures, including clearly stating the way LLM tools were used in the research process, ensuring that human oversight was utilized in the final output, including fact-checking and utilizing as many open-source AI models as possible to ensure reproducibility of information [40].”
MethodShaping generative AI governance in higher education: Insights from student perception
Barus, O. P., Hidayanto, A. N., Handri, E. Y., Sensuse, D. I., & Yaiprasert, C. (2025) · International Journal of Educational Research Open · doi:10.1016/j.ijedro.2025.100452 · Cited by 93 · see citing works
Πέλλας, Ν. (2023) · Smart Learning Environments · doi:10.1186/s40561-023-00276-4 · Cited by 87 · see citing works
“Given the growing significance of video content in online consumer experiences, it is crucial to comprehend how generative AI will reshape online video platforms and the creative industries at large (Fuchs, 2023; Meyer et al., 2023).”
BackgroundCurrent status of ChatGPT use in medical education: potentials, challenges, and strategies
Xu, T., Weng, H., Liu, F., Yang, L., Luo, Y., Ding, Z., & Wang, Q. (2024) · Journal of Medical Internet Research · doi:10.2196/57896 · Cited by 84 · see citing works
“Encouraging students to preserve transcripts of their interactions using ChatGPT can also help [31].”
Naeem, M., Smith, T., & Thomas, L. (2025) · International Journal of Qualitative Methods · doi:10.1177/16094069251333886 · Cited by 126 · see citing works
“The use of AI in qualitative research expedites the data analytical process (Prescott et al., 2024), but it also raises issues of conceptuality, ethics and transparency (Meyer et al., 2023; Salimi & Saheb, 2023).”
Artificial Intelligence Integration: Pedagogical Strategies and Policies at Leading Universities
Alqahtani, N., & Wafula, Z. (2025) · Innovative Higher Education · doi:10.1007/s10755-024-09749-x · Cited by 84 · see citing works
“AI Large Language models, such as ChatGPT, Bard, and Claude, have enhanced the educator’s pedagogical productivity (Li et al., 2024; Meyer et al., 2023 ).”
Chow, J. C. L., & Li, K. (2024) · JMIR Bioinformatics and Biotechnology · doi:10.2196/64406 · Cited by 75 · see citing works
Pushpanathan, K., Lim, Z. W., Yew, S. M. E., Chen, D. Z., Lin, H. A. H., Goh, J. H. L., Wong, W., Wang, X., Tan, M. C. J., Koh, V., et al. (2023) · iScience · doi:10.1016/j.isci.2023.108163 · Cited by 78 · see citing works
Data extraction from polymer literature using large language models
Gupta, S., Mahmood, A. U., Shetty, P., Adeboye, A., & Ramprasad, R. (2024) · Communications Materials · doi:10.1038/s43246-024-00708-9 · Cited by 78 · see citing works
“However, obtaining the desired output from an LLM poses a challenging task and is presently a subject of active research 33,34 .”
Large language models for life cycle assessments: Opportunities, challenges, and risks
Preuss, N., Alshehri, A. S., & You, F. (2024) · Journal of Cleaner Production · doi:10.1016/j.jclepro.2024.142824 · Cited by 71 · see citing works
Wang, F., Li, N., Cheung, A. C. K., & Wong, G. K. W. (2025) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-025-00547-9 · Cited by 69 · see citing works
“…importance of understanding learners’ trust in GenAI should not be overstated in the current era of GenAI, especially when a growing body of research is highlighting students’ disuse, misuse, and abuse of GenAI in education (e.g., Klingbeil et al., 2024; Meyer et al., 2023; Shahzad et al., 2024).”
InfluentialPudasaini, S., Miralles-Pechuán, L., Lillis, D., & Llorens Salvador, M. (2025) · Journal of Academic Ethics · doi:10.1007/s10805-024-09576-x · Cited by 70 · see citing works
“One major concern is determining authorship and intellectual property rights for texts generated with LLMs, particularly in collaborative environments where LLMs might contribute partially (Meyer et al., 2023).”
Generative AI and simulation modeling: how should you (not) use large language models like ChatGPT
Akhavan, A., & Jalali, M. S. (2024) · System Dynamics Review · doi:10.1002/sdr.1773 · Cited by 68 · see citing works
Introduction to Artificial Intelligence
Papajorgji, P. J., & Moskowitz, H. (2024) · Springer eBooks · doi:10.1007/978-3-031-78078-3_5 · Cited by 65 · see citing works
Kooli, C., & Yusuf, N. (2024) · International Journal of Human–Computer Interaction · doi:10.1080/10447318.2024.2338330 · Cited by 63 · see citing works
“For these reasons, the decision to integrate or restrict AI technologies like ChatGPT in educational settings should be based on a well-informed scholarly consensus that would consider evidence bases from various fields (Kooli, 2023b; Meyer et al., 2023).”
Symeou, L., Louca, L. Τ., Kavadella, A., Mackay, J., Danidou, Y., & Raffay, V. I. (2025) · European Journal Of Dental Education · doi:10.1111/eje.13069 · Cited by 82 · see citing works
A personal health large language model for sleep and fitness coaching
Khasentino, J., Belyaeva, A., Liu, X., Yang, Z., Furlotte, N. A., Lee, C., Schenck, E., Patel, Y., Cui, J. H., Schneider, L. D., et al. (2025) · Nature Medicine · doi:10.1038/s41591-025-03888-0 · Cited by 104 · see citing works
“Few-shot prompting with self-consistency improved performance further in both sleep and fitness examinations (Supplementary Table 6).”
Analyzing student prompts and their effect on ChatGPT’s performance
Sawalha, G., Taj, I., & Shoufan, A. (2024) · Cogent Education · doi:10.1080/2331186x.2024.2397200 · Cited by 74 · see citing works
“However, the acceptance of LLMs in education - particularly in Engineering and computation - will depend on their accuracy and response correctness (Meyer et al., 2023), which in turn relies on the quality of the prompt (Knoth et al., 2024; Stojanov et al., 2024; Wang et al., 2024; White et al.,…”
Idris, M. D., Feng, X., & Dyo, V. (2024) · IEEE Access · doi:10.1109/access.2024.3400164 · Cited by 57 · see citing works
Zhu, S., Wang, Z., Zhuang, Y., Jiang, Y., Guo, M., Zhang, X., & Gao, Z. (2024) · Telematics and Informatics Reports · doi:10.1016/j.teler.2024.100138 · Cited by 55 · see citing works
Comparing GPT-4 and Human Researchers in Health Care Data Analysis: Qualitative Description Study
Li, K. D., Fernandez, A., Schwartz, R., Ríos, N., Carlisle, M., Amend, G., Patel, H. V., & Breyer, B. N. (2024) · Journal of Medical Internet Research · doi:10.2196/56500 · Cited by 55 · see citing works
“These developments raise critical questions about their potential and ethical use [1-3].”
Cross-disciplinary perspectives on the potential for artificial intelligence across chemistry
Mroz, A. M., Basford, A. R., Hastedt, F., Jayasekera, I. S., Mosquera‐Lois, I., Sedgwick, R., Ballester, P. J., Bocarsly, J. D., del Rio‐Chanona, E. A., Evans, M. L., et al. (2025) · Chemical Society Reviews · doi:10.1039/d5cs00146c · Cited by 59 · see citing works
Rane, N. L. (2024) · International Journal of Artificial Intelligence and Machine Learning · doi:10.51483/ijaiml.4.1.2024.22-47 · Cited by 53 · see citing works
Evolution of Learning: Assessing the Transformative Impact of Generative AI on Higher Education
Krause, S., Panchal, B. H., & Ubhe, N. (2025) · Frontiers of digital education. · doi:10.1007/s44366-025-0058-7 · Cited by 70 · see citing works
Zhou, J., Zhang, J., Wan, R., Cui, X., Liu, Q., Guo, H., Shi, X., Fu, B., Meng, J., Yue, B., et al. (2025) · BMC Medical Education · doi:10.1186/s12909-025-06916-2 · Cited by 55 · see citing works
Watson, S., & Romic, J. (2025) · European Educational Research Journal · doi:10.1177/14749041231221266 · Cited by 50 · see citing works
“This approach mitigates extreme utopian or dystopian perspectives, guiding a more nuanced discourse around the responsible integration of LLMs into educational and social systems (Meyer et al., 2023).”
Ma, Y., Lan, Y., Xie, Y., Yu, L., Chen, C., Wu, Y., & Dai, X. (2024) · Remote Sensing · doi:10.3390/rs16020404 · Cited by 45 · see citing works
“BERT [38] and GPT [39], among other NLP models, have demonstrated the powerful learning capabilities of transformer, which has also attracted researchers to explore the application of transformer in computer vision tasks.”
Bjelobaba, S., Waddington, L., Perkins, M., Foltýnek, T., Bhattacharyya, S., & Weber-Wulff, D. (2025) · International Journal for Educational Integrity · doi:10.1007/s40979-025-00191-w · Cited by 96 · see citing works
“The integration of GenAI tools into this complex writing process requires careful consideration and potentially new guidelines from funding agencies (Meyer et al. 2023) to ensure research integrity.”
Artificial intelligence-powered tools and academic writing: to use or not to use ChatGPT
Al-Sofi, B. (2024) · Saudi Journal of Language Studies · doi:10.1108/sjls-06-2024-0029 · Cited by 45 · see citing works
Harnessing generative artificial intelligence to support nature‐based solutions
Richards, D. R., Worden, D., Song, X. P., & Lavorel, S. (2024) · People and Nature · doi:10.1002/pan3.10622 · Cited by 44 · see citing works
Faisal, E. (2024) · Frontiers in Education · doi:10.3389/feduc.2024.1325601 · Cited by 44 · see citing works
Pereira, E., Nsair, S., Pereira, L. R., & Grant, K. A. (2024) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-024-00457-2 · Cited by 63 · see citing works
“Furthermore, given the risks associated with the misuse of AI (IEEE, n.d.) as well as equity, diversity, and inclusion (EDI) concerns (Meyer et al., 2023) associated with LLMs, users should be aware that the output from these models can be biased and inaccurate.”
BackgroundApplications of ChatGPT in Otolaryngology–Head Neck Surgery: A State of the Art Review
Lechien, J. R., & Rameau, A. (2024) · Otolaryngology · doi:10.1002/ohn.807 · Cited by 42 · see citing works
Sinha, R., Solola, I., Nguyen, H., Swanson, H., & Lawrence, L. (2024) · Symposium on Learning, Design and Technology (LDT) · doi:10.1145/3663433.3663456 · Cited by 42 · see citing works
Impact of Large Language Models on Medical Education and Teaching Adaptations
Zhui, L., Yhap, N., Liping, L., Wang, Z., Zhonghao, X., Xiaoshu, Y., Hong, C., Liu, X., & Wei, R. (2024) · JMIR Medical Informatics · doi:10.2196/55933 · Cited by 49 · see citing works
“This analytical capability fosters a more nuanced understanding of student performance, thus enabling educators to tailor their teaching strategies effectively [25].”
BackgroundDo AI-generative tools kill or nurture creativity in EFL teaching and learning?
Alzubi, A. A. F., Nazim, M., & Alyami, N. (2025) · Education and Information Technologies · doi:10.1007/s10639-025-13409-8 · Cited by 43 · see citing works
“Kohnke et al. ( 2023 ) and Meyer et al. ( 2023 ) also emphasized that, despite their abil - ity to generate creative content, AI tools cannot replace the rich, interactive nature of human collaboration, which is vital for deeper creativity.”
Busuttil, L., & Calleja, J. (2025) · Digital Experiences in Mathematics Education · doi:10.1007/s40751-024-00168-3 · Cited by 59 · see citing works
“Our findings resonate with studies highlighting ChatGPT’s potential in personalising learning and providing immediate feedback, aligning with Memarian and Doleck (2023) and Meyer et al. (2023), who stress the significance of individualised learning trajectories and interactive revision aids.”
InfluentialChatGPT as an AI L2 teaching support: A case study of an EFL teacher
Octavio, M. M., Argüello, M. V. G., & Pujolà, J. (2024) · Technology in Language Teaching & Learning · doi:10.29140/tltl.v6n1.1142 · Cited by 39 · see citing works
Creative Applications of Artificial Intelligence in Education
Urmeneta, A., & Roméro, M. (2024) · Palgrave studies in creativity and culture · doi:10.1007/978-3-031-55272-4 · Cited by 52 · see citing works
Bhattacharya, M., Pal, S., Chatterjee, S., Lee, S., & Chakraborty, C. (2024) · Molecular Therapy — Nucleic Acids · doi:10.1016/j.omtn.2024.102255 · Cited by 38 · see citing works
Human versus machine: The effectiveness of ChatGPT in automated essay scoring
Manning, J., Baldwin, J., & Powell, N. (2025) · Innovations in Education and Teaching International · doi:10.1080/14703297.2025.2469089 · Cited by 36 · see citing works
“Additionally, educators must be vigilant regarding the potential ethical implications of using AI in grading, including the transparency of AI systems, the possibility of bias, and the fairness of grading practices (Meyer et al., 2023).”
Al-Abri, A. (2025) · Education and Information Technologies · doi:10.1007/s10639-025-13484-x · Cited by 44 · see citing works
“The inclusion of practice quizzes and detailed explanations supports the development of self-regulated learning strategies, and reduces exam-related anxiety as students better gauge their preparation levels (Cheung et al., 2023; Meyer et al., 2023).”
A Practical Guide and Assessment on Using ChatGPT to Conduct Grounded Theory: Tutorial
Yue, Y., Liu, D., Lv, Y., Hao, J., & Cui, P. (2025) · Journal of Medical Internet Research · doi:10.2196/70122 · Cited by 35 · see citing works
Uppal, K., & Hajian, S. (2024) · European Journal of Educational Research · doi:10.12973/eu-jer.14.1.199 · Cited by 36 · see citing works
Large language models and medical education: a paradigm shift in educator roles
Zhui, L., Li, F., Fu, Q., Wang, X., Liu, H., Zhao, Y., & Wei, R. (2024) · Smart Learning Environments · doi:10.1186/s40561-024-00313-w · Cited by 36 · see citing works
“The employment of LLMs streamlines the exam preparation process, conserves valuable time, and, by analysing students’ learning data and examination results, aids educators in discerning students’ learning patterns and challenges to refine teaching strategies (Meyer et al., 2023).”
BackgroundPirzado, F. A., Ahmed, A., Mendoza-Urdiales, R. A., & Terashima‐Marín, H. (2024) · IEEE Access · doi:10.1109/access.2024.3443621 · Cited by 37 · see citing works
“These findings align with some of the studies included in this review [28], [62].”
ResultConstruction of Cultural Heritage Knowledge Graph Based on Graph Attention Neural Network
Wang, Y., Liu, J., Wang, W., Chen, J., Yang, X., Sang, L., Wen, Z., & Peng, Q. (2024) · Applied Sciences · doi:10.3390/app14188231 · Cited by 34 · see citing works
Anson, D. W. J. (2024) · Higher Education Research & Development · doi:10.1080/07294360.2024.2332259 · Cited by 41 · see citing works
Zhu, Y., Zhu, C., Wu, T., Wang, S., Zhou, Y., Chen, J., Wu, F., & Li, Y. (2025) · Education and Information Technologies · doi:10.1007/s10639-024-12898-3 · Cited by 33 · see citing works
“They have the potential to revolutionize various facets of human daily life, including education (Meyer et al., 2023; Tamkin et al., 2021; Rillig et al., 2023).”
InfluentialKavitha, K., Joshith, V. P., & Sharma, S. (2024) · E-Learning and Digital Media · doi:10.1177/20427530241259099 · Cited by 32 · see citing works
“…described a wide range of assistance, including guidance on managing tasks, generating assignment ideas, explaining concepts in depth, offering new learnings, and providing systematic guidance on completing important tasks to enhance productivity (Javaid et al., 2023b; Meyer et al., 2023).”
Integrating vision‐based AI and large language models for real‐time water pollution surveillance
Samuel, R. D. J., Sermet, Y., Cwiertny, D. M., & Demir, İ. (2024) · Water Environment Research · doi:10.1002/wer.11092 · Cited by 36 · see citing works
“Meyer et al. ( 2023) investigates the potential use of LLM based chatbots in academic works and ethical implication (Sajja, Sermet, Cwiertny, et al., 2023b).”
Lin, Y., Luo, Z., Ye, Z., Zhong, N., Zhao, L., Zhang, L., Li, X., Chen, Z., & Chen, Y. (2025) · JMIR Medical Education · doi:10.2196/71125 · Cited by 31 · see citing works
“…a critical technological imbalance: the overwhelming focus on general-purpose LLMs like ChatGPT contrasts sharply with the lack of systematic development of specialized medical models and the near absence of research on multimodel collaborative mechanisms within medical education [10,164].”
What Does ChatGPT Mean for Qualitative Health Research ?
van Manen, M. (2023) · Qualitative Health Research · doi:10.1177/10497323231210816 · Cited by 31 · see citing works
“…2023; Baker et al., 2023; Cohen et al., 2023; Conroy et al., 2023; Dergaa et al., 2023; Dien, 2023; Doshi et al., 2023; Doyal et al., 2023; Farina & Lavazza, 2023; Garcia, 2023; Jeyaraman et al., 2023; Leung et al., 2023; Meyer et al., 2023; Ruf fl e et al., 2023; Takeda, 2023; Wang et al., 2023).”
Popkov, A. A., & Barrett, T. S. (2024) · Accountability in Research · doi:10.1080/08989621.2024.2331757 · Cited by 35 · see citing works
“Similarly, generative AI models like ChatGPT now provide a new level of support in writing but also cause new debates about academic authorship and integrity (DuBose and Marshall 2023; Else 2023; Flanagin et al. 2023; Meyer et al. 2023).”
Prandner, D., Wetzelhütter, D., & Hese, S. (2025) · Frontiers in Education · doi:10.3389/feduc.2024.1417900 · Cited by 43 · see citing works
Fontoura, L., Nascimento, D. L. D. M., Neto, J. V., & Caiado, R. G. G. (2025) · Technology in Society · doi:10.1016/j.techsoc.2025.102847 · Cited by 41 · see citing works
“…and bias mitigation, which studies caution against the unscrupulous and irresponsible design of Gen-AI technologies, as such practices may perpetuate systemic prejudices and disparities, particularly affecting marginalised groups and potentially exacerbating existing biases (Meyer et al., 2023).”
Alyasiri, O. M., Salisu, S., Younis, H., Salman, A. M., Sahib, T. M., Akhtom, D., & Hayder, I. M. (2024) · Journal of Stomatology Oral and Maxillofacial Surgery · doi:10.1016/j.jormas.2024.101842 · Cited by 31 · see citing works
Huo, B., Calabrese, E., Sylla, P., Kumar, S., Ignacio, R. C., Oviedo, R., Hassan, I., Slater, B. J., Kaiser, A., Walsh, D. S., et al. (2024) · Surgical Endoscopy · doi:10.1007/s00464-024-10807-w · Cited by 28 · see citing works
Luo, C., Mao, B., Wu, Y., & He, Y. (2024) · Nurse Education Today · doi:10.1016/j.nedt.2024.106321 · Cited by 32 · see citing works
Large Language Models in Pediatric Education: Current Uses and Future Potential
Suresh, S., & Misra, S. M. (2024) · PEDIATRICS · doi:10.1542/peds.2023-064683 · Cited by 29 · see citing works
Global retractions due to randomly generated content: Characterization and trends
Lei, F., Du, L., Dong, M., & Liu, X. (2024) · Scientometrics · doi:10.1007/s11192-024-05172-3 · Cited by 35 · see citing works
“Nevertheless, the fundamental challenge is that they can present false information as truth, also known as “illusions” (Meyer et al., 2023).”
Hassan, A., Nizam‐Uddin, N., Quddus, A., Hassan, S. R., Rehman, A. U., & Bharany, S. (2024) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2024.057877 · Cited by 28 · see citing works
Prompt engineering: The next big skill in rheumatology research
Venerito, V., Lalwani, D., Vescovo, S. D., Iannone, F., & Gupta, L. (2024) · International Journal of Rheumatic Diseases · doi:10.1111/1756-185x.15157 · Cited by 27 · see citing works
ChatGPT: a transformative role in academia–insights into academic staff performance since adoption
Sadallah, M., Bin‐Nashwan, S. A., & Benlahcene, A. (2024) · Journal of Information Communication and Ethics in Society · doi:10.1108/jices-07-2024-0097 · Cited by 30 · see citing works
Alomari, E. A. (2024) · Computer Modeling in Engineering & Sciences · doi:10.32604/cmes.2024.052256 · Cited by 32 · see citing works
Mathebula, M., Modupe, A., & Marivate, V. N. (2024) · IEEE Access · doi:10.1109/access.2024.3464374 · Cited by 29 · see citing works
“ChatGPT is one of the largest language models and contains the most parameters, with 175 billion [57].”
BackgroundSaleh, Y., Talib, M. W. A., Nasir, Q., & Dakalbab, F. M. (2025) · Frontiers in Computer Science · doi:10.3389/fcomp.2025.1523699 · Cited by 41 · see citing works
Machine Learning for Organic Fluorescent Materials
Zhong, J., Zhu, W., Shen, S., Zhou, N., Xi, M., Du, K., Wang, D., & Tang, B. Z. (2025) · Aggregate · doi:10.1002/agt2.70089 · Cited by 26 · see citing works
“Although data collection can be tedious, the rise of artificial intelligence, particularly Large Language Models (LLMs) [39], has facilitated the use of automated tools for data mining.”
A Systematic Literature Review of Empirical Research on ChatGPT in EducationPreprint
Albadarin, Y., Tukiainen, M., Saqr, M., & Pope, N. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4562771 · Cited by 26 · see citing works
Kortemeyer, G., & Nöhl, J. (2025) · Physical Review Physics Education Research · doi:10.1103/physrevphyseducres.21.010136 · Cited by 39 · see citing works
“Most recently, Large Language Models (LLMs), in particular generative Pre-trained Transformers (GPT) [1], such as GPT-4 [2], have demonstrated impressive abilities in solving and evaluating academic tasks [3], including tasks related to academic education [4].”
Teaching and testing in the era of text-generative AI: exploring the needs of students and teachers
Jochim, J., & Lenz-Kesekamp, V. (2024) · Information and Learning Sciences · doi:10.1108/ils-10-2023-0165 · Cited by 24 · see citing works
LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing
Du, J., Wang, Y., Zhao, W., Deng, Z., Liu, S., Lou, R., Zou, H. P., Narayanan Venkit, P., Zhang, N., Srinath, M., et al. (2024) · Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2024.emnlp-main.292 · Cited by 23 · see citing works
Schnepper, R., Roemmel, N., Schaefert, R., Lambrecht-Walzinger, L., & Meinlschmidt, G. (2024) · JMIR Mental Health · doi:10.2196/57986 · Cited by 23 · see citing works
“This may not only facilitate the automatic analysis of large amounts of text data [9,10] but, by modeling words in a large context, also allow the generation of meaningful text and the interactive use of this technology [5,10].”
Ballard, D. H., Antigua-Made, A., Barré, E., Edney, E. A., Gordon, E. B., Kelahan, L. C., Lodhi, T., Martin, J. G., Özkan, M. U., Serdynski, K. C., et al. (2024) · Academic Radiology · doi:10.1016/j.acra.2024.10.023 · Cited by 23 · see citing works
Aloudat, M. Z., Aboumadi, A., Soliman, A., Al-Mohammed, H. A., Al-Ali, M. S., Mahgoub, A., Barhamgi, M., & Yaacoub, E. E. (2025) · IEEE Access · doi:10.1109/access.2025.3555753 · Cited by 27 · see citing works
“This involves speaking in a way appropriate for the characters, setting, and intended tone—all of which might change greatly throughout virtual worlds [175], [252]. d) Scalability: Scalability is an issue for LLMs in the metaverse, just like it is with blockchain [253], [254].”
Gao, Y., Zhai, X.-M., Li, M., Lee, G., & Liu, X.-X. (2025) · Journal of Research in Science Teaching · doi:10.1002/tea.70009 · Cited by 22 · see citing works
Jamshed, M., Ahmed, A. S. M. M., Sarfaraj, M., & Warda, W. U. (2024) · International Journal of Interactive Mobile Technologies (iJIM) · doi:10.3991/ijim.v18i19.50361 · Cited by 24 · see citing works
Wang, J., Hu, H., Wang, Z., Yan, S., Sheng, Y., & He, D. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3641917 · Cited by 21 · see citing works
- Authors
- Meyer, J. G., Urbanowicz, R. J., Martin, P. C. N., O’Connor, K., Li, R., Peng, P.-C., Bright, T. J., Tatonetti, N., Won, K. J., Gonzalez-Hernandez, G., & Moore, J. H.
- Affiliations
- Cedars-Sinai Medical Center (US)
- University of Pennsylvania (US)
- Corresponding author
- Meyer, Jesse G.
- Moore, Jason H.
- ORCID
- Venue
- BioData Mining
- Published
- 2023-07-13
- Volume
- 16
- Issue
- 1
- Pages
- 20-20
- ISSN
- 1756-0381 (ISSN-L)
- PMC
- PMC10339472
- OpenAlex
- W4384200891
- Semantic Scholar
- e95b10b9a8415874f24275b665628fc340eee5dc
- Cited by
- 569
- References
- 23
- Citations by year
- 2026: 104 · 2025: 215 · 2024: 220 · 2023: 29
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- BioData Mining (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- National Institute on Aging (AG066833)
- National Institute of General Medical Sciences (R35GM142502)
- National Institute of Allergy and Infectious Diseases (R01AI164481)
- U.S. National Library of Medicine (R01LM011176)
- Topics
- Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- large language models, ChatGPT, academic writing, artificial intelligence in education, AI-assisted programming, plagiarism, algorithmic bias, AI accuracy, research tools, ethics of AI use
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
