The DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index; none of the 18 sources its claims cite is checked for support yet.
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.
Overview
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
Verification
Claims the paper makes
17 Own interpretation · 9 Not checked yet · 7 No basis given · 4 Own result
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Background
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Background · plain
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“In fact, ChatGPT and Bard can misrepresent their capabilities.”
§ ChatGPT and large language models in academia: opportunities and challenges > Background
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“For example, LLMs may adopt bias, perpetuate stereotypes in the training dataset, and present false information as truth.”
§ ChatGPT and large language models in academia: opportunities and challenges > Introduction
What the sources say
- Not checked yetcites [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 yetcites [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 yetcites [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
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Background
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thus, an answer generated by a large language model can be formatted correctly but not necessarily be factual.”
§ ChatGPT and large language models in academia: opportunities and challenges > Background
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Over reliance may hinder students ability to develop critical skills, such as writing [21].”
§ ChatGPT and large language models in academia: opportunities and challenges > Education
What the sources say
- Not checked yetcites [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.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > ChatGPT for academic writing > ChatGPT as a writing and editing tool
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“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].”
§ ChatGPT and large language models in academia: opportunities and challenges > Bias
What the sources say
- Not checked yetcites [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 yetcites [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 yetcites [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
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“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].”
§ ChatGPT and large language models in academia: opportunities and challenges > Scalability and accessibility of LLMs
What the sources say
- Not checked yetcites [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
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“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].”
§ ChatGPT and large language models in academia: opportunities and challenges > Bias
What the sources say
- Not checked yetcites [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 yetcites [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
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: unclear — It is unclear whether it could be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Probably not a substantial claim. 23
- It may state two separate claims as one. 11
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > ChatGPT for academic writing > ChatGPT and LLMs for grant proposal writing
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > ChatGPT for academic writing > ChatGPT and LLMs for grant proposal writing
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“ChatGPT can even provide general text that might be logically used in parts of the grant, such as the aims page.”
§ ChatGPT and large language models in academia: opportunities and challenges > ChatGPT for academic writing > ChatGPT and LLMs for grant proposal writing
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Education
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can suggest specific experiments that might test a hypothesis of interest.”
§ ChatGPT and large language models in academia: opportunities and challenges > ChatGPT for academic writing > ChatGPT and LLMs for grant proposal writing
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > ChatGPT for academic writing > ChatGPT and LLMs for grant proposal writing
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The model can generate correctly formatted information such as references, that do not exist.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“ChatGPT will misrepresent its knowledge; instead of responding “I don’t know”, it will readily provide fabricated information with apparent confidence.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“For example, LLMs may adopt bias, perpetuate stereotypes in the training dataset, and present false information as truth.”
§ ChatGPT and large language models in academia: opportunities and challenges > Introduction
What the sources say
- Not checked yetcites [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 yetcites [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 yetcites [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
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Background
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Scalability and accessibility of LLMs
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In line with warnings for general use, ChatGPT may somewhat unpredictably present incorrect code as being correct.”
§ ChatGPT and large language models in academia: opportunities and challenges > Assistance with programming
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
In the paper
“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].”
§ ChatGPT and large language models in academia: opportunities and challenges > Education
What the sources say
- Not checked yetcites [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
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Assistance with programming
hedgedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Assistance with programming
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Assistance with programming
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“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].”
§ ChatGPT and large language models in academia: opportunities and challenges > Education
What the sources say
- Not checked yetcites [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 yetcites [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 yetcites [17] Tate TP, Doroudi S, Ritchie D, Xu Y, Mark Warschauer UCI. Educational Research and AI-Generated Writing: Confronting the Coming Tsunami. 2023.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“While AI has the potential to revolutionize education and academic research, paywalled access could further increase the divide between the wealthy and the poor.”
§ ChatGPT and large language models in academia: opportunities and challenges > Scalability and accessibility of LLMs
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
- Rests on an assumption: AI has the potential to revolutionize education and academic research.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Education
What the sources say
- Not checked yetcites [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
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This editorial provides several examples and perspectives on how LLMs increase the efficiency of academic writing, education, and programming.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In many contexts, LLMs have low accuracy.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Output from ChatGPT will match existing text on the internet, particularly if the sample is small enough.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
- Rests on an assumption: The output of ChatGPT is from most publicly available text on the internet until 2021.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“However, ChatGPT does not do anything without a prompt.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“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.”
§ ChatGPT and large language models in academia: opportunities and challenges > Bias
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The performance of ChatGPT and other LLMs therefore mimics the available text, and they are as biased as their training data.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
- 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.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thus, in a strict sense, LLMs like ChatGPT do produce plagiarism.”
§ ChatGPT and large language models in academia: opportunities and challenges > Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
- Rests on an assumption: The output of ChatGPT is from most publicly available text on the internet until 2021.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As companies race to innovate with LLMs like ChatGPT and others, encoded biases will be amplified, and harm will manifest.”
§ ChatGPT and large language models in academia: opportunities and challenges > Bias
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds. It may state two separate claims as one.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“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
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“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
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“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
In the citing paper
“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]
In the citing paper
“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]
In the citing paper
“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.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ ChatGPT and large language models in academia: opportunities and challenges > ChatGPT for academic writing > Ethics and rules of using text from ChatGPT
“In one case, ChatGPT was listed as an author in the preprint [9] but missing as an author in the final publication [4].”
§ ChatGPT and large language models in academia: opportunities and challenges > References
“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”
Citations
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) · doi:10.35542/osf.io/5er8f · Cited by 7,552
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×
“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×
“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
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×
“In one case, ChatGPT was listed as an author in the preprint [9] but missing as an author in the final publication [4].”
BackgroundBAİDOO-ANU, D., & OWUSU ANSAH, L. (2023) · Journal of AI · doi:10.61969/jai.1337500 · Cited by 3,939
Sallam, M. (2023) · Healthcare · doi:10.3390/healthcare11060887 · Cited by 3,126 · cited 1×
“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
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
Baidoo-Anu, D., & Owusu Ansah, L. (2023) · doi:10.2139/ssrn.4337484 · Cited by 3,004 · cited 1×
“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.”
Background
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
Items 301–400 of 574
Cui, H., Zhang, Z., Zhang, J., Du, C., Zhang, W., & Cui, H. (2025) · Aesthetic Plastic Surgery · doi:10.1007/s00266-025-05393-8 · Cited by 3
“However, despite the growing applications of LLMs, concerns remain regarding the accuracy, safety, and ethical considerations of their recommendations in various medical fields [4–6].”
Understanding Online Discussion Experiences of Blind Screen Reader Users
Ferdous, M. J., Nayak, A. K., Prakash, Y., Venkatraman, N., Jayarathna, S., Lee, H.-N., & Ashok, V. G. (2025) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2025.2589865 · Cited by 4
Portuguese secondary school students’ perceptions regarding the use of ChatGPT
Callado, A., Saúde, I., & Araújo, J. L. (2025) · Contemporary Educational Technology · doi:10.30935/cedtech/17516 · Cited by 9
Chen, Z., Wang, J., Li, Y., Li, H., Shi, C., Zhang, R., & Qu, H. (2025) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/3746059.3747670 · Cited by 8
Liang, Y., Zhang, J., Chen, J., Wu, J., Shen, H., & Liang, W. (2025) · Ethics & Behavior · doi:10.1080/10508422.2025.2608754 · Cited by 5
“Generative large language models (LLMs) have profoundly changed academic writing practices in higher education (Bao et al., 2025; Meyer et al., 2023; Mo & Crosthwaite, 2025).”
Bohler, F., Tran, T., Burmeister, J. R., Hadid, K., Joshi, S., Gao, S., & Chaiyasate, K. (2025) · Journal of Craniofacial Surgery · doi:10.1097/scs.0000000000012366 · Cited by 4
Kelley, C., Kelley, S., Rivas‐Koehl, M., Wilson, A., Rosinsky, K., Briggs, S., & Scott, M. E. (2026) · Prevention Science · doi:10.1007/s11121-026-01885-4 · Cited by 3
“However, concerns remain about potential inaccuracies (including hallucinations, wherein AI makes false claims), authorship, and intellectual property (Meyer et al., 2023).”
Ethics in publishing: From plagiarism to artificial intelligence
Alnaimat, F., AlSamhori, A. R. F., Seiil, B., Qumar, A. B., & Zimba, O. (2026) · Autoimmunity Reviews · doi:10.1016/j.autrev.2026.104015 · Cited by 7
Sears, L. M., Köseoğlu, M., Antonopoulou, S., Lee, S. K. Y., Kase, M. T., Colebeck, A. C., Lin, Y. (., Huber, H. M., Krzesinski, M., Wee, A. G., et al. (2026) · Journal of Prosthodontics · doi:10.1111/jopr.70136 · Cited by 3
A Framework for Explainable Multi-purpose Virtual Assistants: A Nutrition-Focused Case Study
Buzcu, B., Pannatier, Y., Aydoğan, R., Schumacher, M., Calbimonte, J.-P., & Calvaresi, D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-70074-3_4 · Cited by 3
Generative Artificial Intelligence and Postsecondary Education
Gierhart, A. R., Shefferly, N., Li, Y., & Speetzen, E. D. (2024) · Advances in educational technologies and instructional design book series · doi:10.4018/979-8-3693-5633-3.ch002 · Cited by 4
Farr, D. T., Ng, L. H. X., Cruickshank, I. J., Manzonelli, N., Clark, N., Starbird, K., Bastian, N. D., & West, J. D. (2025) · IEEE Access · doi:10.1109/access.2025.3585414 · Cited by 3
“Large Language Models (LLMs) have transformed artificial intelligent applications, offering powerful tools for processing and analyzing text across diverse domains, including healthcare [1], academia [2], cybersecurity [3], and software development [4].”
Fan, Y., Li, Y., Natalino, C., Guo, J., Wu, W., Ruan, R., Wang, W. M., Zhao, Y., & Zhang, J. (2025) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2025.3590011 · Cited by 3
“For example, large models such as GPT-4 require a massive infrastructure of 25,000 NVIDIA A100 cards [2].”
User Invented Applications of AI-Based GPTs in Teaching-Learning and Research-Publications
Aithal, P. S., & Aithal, S. (2024) · International Journal of Applied Engineering and Management Letters · doi:10.47992/ijaeml.2581.7000.0211 · Cited by 3
EQUIPPING STUDENTS FOR A DYNAMIC FUTURE
Ramamonjisoa, D. (2024) · PUPIL International Journal of Teaching Education and Learning · doi:10.20319/pijtel.2024.82.3248 · Cited by 3
ChatGPT's Performance on Problem Sets in an At-Scale Introductory Computer Science Course
Popescu, D. M., & Joyner, D. A. (2024) · ACM Conference on Learning @ Scale (L@S) · doi:10.1145/3657604.3664701 · Cited by 3
Deriving inferences through natural language from structured datasets for asset lifecycle management
Sanchit, Bhattacharjee, S., & Pandhare, V. (2024) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2024.08.064 · Cited by 3
Navigating Ethical Dilemmas Of Generative AI In Medical Writing
Hamdan, Q. U., Umar, W., & Hasan, M. K. C. (2024) · Journal of Rawalpindi Medical College · doi:10.37939/jrmc.v28i3.2744 · Cited by 3
Harnessing Artificial Intelligence to Transform Education: Challenges and Opportunities
Ahma, G., & Kadriu, A. (2025) · SEEU Review · doi:10.2478/seeur-2025-0020 · Cited by 3
Toward a woven literature: Open-source infrastructure for reproducible publishingPreprint
Karakuzu, A. (2025) · doi:10.55458/neurolibre.00041 · Cited by 3
LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing
Du, J., Wang, Y., Zhao, W., Deng, Z., Liu, S., Lou, R., Zou, H. P., Venkit, P. N., Zhang, N., Srinath, M., et al. (2024) · arXiv · doi:10.48550/arxiv.2406.16253 · Cited by 2
Navigating the Risks: A Review of Safety Issues in Large Language Models
Wang, H., Li, Y., Wang, Y., Liu, P., & Li, P. (2024) · IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C) · doi:10.1109/qrs-c63300.2024.00020 · Cited by 2
“…generated by LLMs could introduce biases into neural retrieval systems and even affect the impartiality of code produced by LLMs. Jesse et al. [27] have found that AI/ML solutions designed without ethical considerations in healthcare settings can intensify systemic biases and disparities,…”
BackgroundPopescu, D. M., & Joyner, D. A. (2024) · IEEE Digital Education and MOOCS Conference (DEMOcon) · doi:10.1109/democon63027.2024.10748200 · Cited by 3
Understanding Optimal Interactions Between Students and A Chatbot During A Programming Task
Shin, J., Castro, L. M. C., Yang, Z., Castelblanco, G., Aggarwal, A., Leite, W. L., & Carroll, B. F. (2024) · Winter Simulation Conference · doi:10.1109/wsc63780.2024.10838799 · Cited by 4
“…the learning process through meaningful and interactive exchanges by optimizing and personalizing educational experiences in higher education (Meyer et al. 2023) Large Language Models (LLMs) are currently being introduced into higher education through various pilot programs and research…”
The Impact of Generative AI on Learning Across Grades
Xie, M., & Luo, L. (2025) · International Conference on Educational and Information Technology · doi:10.1109/iceit64364.2025.10976022 · Cited by 2
“Educators use AI to enhance teaching quality[10][11], while students benefit from AI-powered learning technologies that offer personalized learning paths, real-time feedback, and study recommendations [12].”
Exploring Occupational Gender Bias in Large Language Models of High-Resource Language
Jia, K., Liu, Z., Zhang, D., Chen, H., Luo, F., & Qi, J. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-4506-0_16 · Cited by 2
The Impact of Chatbots on Students' Reflective Thinking in Introductory Programming Course
Rachmat, A., Watterson, C. A., & Lundqvist, K. Ø. (2025) · IEEE Global Engineering Education Conference (EDUCON) · doi:10.1109/educon62633.2025.11016417 · Cited by 7
“Prompting chatbots can help students utilize them for learning [15].”
Zhou, Y., Kang, J., & Nguyen, H. (2025) · Proceedings. · doi:10.22318/icls2025.704957 · Cited by 3
BCN: Enhanced Backpressure Flow Control with Rapid Notification in Datacenter Networks
Hu, X., Hu, D., Dong, D., Li, C., Zhou, Z., & Yuan, G. (2025) · Computer Networks · doi:10.1016/j.comnet.2025.111426 · Cited by 2
AcaQAS: An Academic Question Answering System Based on Finetuning Large Language Models
Lê, T. H., & Duong, B. T. (2025) · SN Computer Science · doi:10.1007/s42979-025-04213-9 · Cited by 2
“Meyer et al. [21] and Ji et al. [22] examined these challenges, while Zhou et al. [23] proposed validation frameworks to improve factual consistency.”
Artificial worlds and artificial minds: Authenticity and language learning in digital lifeworlds
Matthews, B. (2025) · Journal of Responsible Technology · doi:10.1016/j.jrt.2025.100131 · Cited by 2
Zhang, L., Song, X., Li, P., Yuan, Y., Wan, K., Huang, F., Guo, Y., & Zhang, H. (2025) · Arabian Journal of Chemistry · doi:10.25259/ajc_33_2025 · Cited by 2
Mekterović, I., & Brkić, L. (2025) · International Convention on Information and Communication Technology, Electronics and Microelectronics · doi:10.1109/mipro65660.2025.11131962 · Cited by 2
Adeyemi, I. O., Giwa, K., Giwa, F., Hassan, R., & Gabriel, A. (2025) · Artificial Intelligence in Education · doi:10.1108/aiie-11-2024-0041 · Cited by 4
Toward a woven literature: Open-source infrastructure for networked scientific publishing
Karakuzu, A. (2025) · Aperture Neuro · doi:10.52294/001c.146072 · Cited by 2
P, A. M., Dileep Kumar, M., Sharma, V., Tripathi, R., & Badjatia, A. (2025) · Information Discovery and Delivery · doi:10.1108/idd-05-2025-0111 · Cited by 2
The Missing Target: Why Industrialized Animal Farming Must Be at the Core of the Climate Agenda
Mace, J. L., Knight, A., Vieira, F., Tatemoto, P., & Gameiro, M. B. P. (2025) · Animals · doi:10.3390/ani15223256 · Cited by 2
Machine learning-based research of AI marketing: topic analysis and model construction
Xie, H., Nie, Y., Liu, L., & Yang, C. (2025) · Future Business Journal · doi:10.1186/s43093-025-00686-5 · Cited by 7
Yaacoub, J.-P. A., Noura, H., Salman, O., & Pujolle, G. (2025) · Annals of Telecommunications · doi:10.1007/s12243-025-01134-9 · Cited by 3
Bernik, N., & Šprajc, P. (2025) · Organizacija · doi:10.2478/orga-2025-0024 · Cited by 2
“At the same time, their introduction raises new theoretical questions concerning trust, employee acceptance, and the socio-technical integration of AI into organisational processes (Meyer et al., 2023).”
Personalised Learning for Nursing Education With Gen-AI Chatbots
Mathews, P., Adams, J., & Cheyne, C. (2025) · Nursing Praxis in Aotearoa New Zealand · doi:10.36951/001c.151665 · Cited by 2
Emotional Strain and Frustration in LLM Interactions in Software Engineering
Montes, C. M., & Khojah, R. (2025) · International Conference on Evaluation and Assessment in Software Engineering (EASE) · doi:10.1145/3756681.3756951 · Cited by 3
Marconi, L., Brasca, F., & Ferri, P. M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-13187-4_14 · Cited by 3
Validity and Trustworthiness of AI-Generated Information in an Online Nutrition Course
Briggs, M., Archibald, A., Connors, P., & Heap, T. (2026) · Journal of Hospitality & Tourism Education · doi:10.1080/10963758.2026.2619164 · Cited by 2
Evaluating the Accuracy and Stability of Frontier LLMs on ENADE Computer Science Questions
Carvalho, L. D. M., Carvalho, C. M., & Mendonça, N. C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-15993-9_18 · Cited by 2
Adoption of Generative AI in Higher Education: Perceptions of Journalism Students
Alonso-Muñoz, L., & Casero-Ripollés, A. (2026) · Information · doi:10.3390/info17020189 · Cited by 2
“…the ethical part of Gen AI use by the student body [17,38], preventing them from making an unintentional and erroneous use of these platforms [10,13–15 Finally, regarding how Journalism students wish to be trained on Gen AI (RQ5), the data indicate that the majority would like this type of…”
Al-Shamaileh, O., Hammady, R., Abdelrahman, M., & Mubin, O. (2026) · Computers in Human Behavior Reports · doi:10.1016/j.chbr.2026.100980 · Cited by 6
“…Chen et al., 2024; Farrelly & Baker, 2023; Frick, 2024; Ghorashi et al., 2023; Gruenhagen, Sinclair, et al., 2024; Kayalı et al., 2023; Kooli, 2023; Meyer et al., 2023; Nguyen et al., 2024; Pallivathukal et al., 2024; Phutela et al., 2024; Sijing & Lan, 2018; Zeb et al., 2024) Overreliance &…”
Generative AI as a Knowledge Distribution System
Lavazza, A., & Farina, M. (2026) · Social Epistemology · doi:10.1080/02691728.2026.2623557 · Cited by 4
Wong, J. M. S., Yip, C. W. H., & Liu, P. H. C. (2026) · AAOU Journal/AAOU journal · doi:10.1108/aaouj-10-2025-0159 · Cited by 2
Outsmarting ChatGPT in Chemistry: Using Student-Formulated Questions for Critical AI Evaluation
López-Pérez, K., Benjamin, S.-A., & Veige, M. K. (2026) · Journal of Chemical Education · doi:10.1021/acs.jchemed.5c01611 · Cited by 2
Navigating Supervisory Conversations About Gen AI : Taboo and Supervisory Power Dynamics
Qi, J., Tuxworth, J., & Gomes, C. (2026) · Higher Education Quarterly · doi:10.1111/hequ.70151 · Cited by 2
Laila, U., Shah, Z. A., Yu, P., Mulli, J., Khurram, M., & Ahmed, M. M. (2023) · Advances in higher education and professional development book series · doi:10.4018/979-8-3693-0487-7.ch012 · Cited by 2
ChatGPT and Education: A Scopus Bibliometric Analysis
Riyadi, A. R., Syaripudin, T., Kurniasih, Hendriani, A., Sutinah, C., & Sirojudin, M. A. (2024) · Advances in Social Science, Education and Humanities Research/Advances in social science, education and humanities research · doi:10.2991/978-2-38476-206-4_30 · Cited by 3
Meta-requirements for LLM-Based Knowledge Exploration Tools in Information Systems Research
Sjöström, J., & Cronholm, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-61175-9_29 · Cited by 2
The Use and Effects of ChatGPT in Tourism Education
Yaşar, E., Öztürk, İ., Yayla, E., & TÜR, E. (2024) · Advances in systems analysis, software engineering, and high performance computing book series · doi:10.4018/979-8-3693-3033-3.ch007 · Cited by 2
Tolety, V. B. P., & Evani, V. P. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-0892-5_59 · Cited by 2
Large Language Model in Various Fields: Opportunities, Challenges and Risks
Ghosh, R. C., Shailendra, P., & Singh, G. B. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-3604-1_39 · Cited by 2
Barua, M. (2024) · Cureus · doi:10.7759/cureus.72293 · Cited by 3
MedQAS: A Medical Question Answering System Based on Finetuning Large Language Models
Duong, B. T., & Lê, T. H. (2024) · Communications in computer and information science · doi:10.1007/978-981-96-0434-0_21 · Cited by 2
Understanding Algorithmic Problem Solving using LLMs
Velez, X. (2024) · ACM Virtual Global Computing Education Conference (SIGCSE Virtual) · doi:10.1145/3649409.3691086 · Cited by 2
Prasodjo, P., Kweldju, S., Mukminatien, N., & Ivone, F. M. (2025) · International Journal of Religion · doi:10.61707/hb2kzm10 · Cited by 5
“Within the Artificial Intelligence technology, Large Language Models (LLMs) occupy a unique position (Meyer et al., 2023).”
Use of large language model (LLM) to enhance content and structure of a school of dentistry LibGuide
Jones, E. P. (2025) · Journal of the Medical Library Association JMLA · doi:10.5195/jmla.2025.2084 · Cited by 2
Effective Diversity Optimizations for High Accuracy Deep Ensembles
Jin, H., Aghdam, M. A., Medikonduru, S. N. C., Wei, W., Wang, X., Zhang, W., & Wu, Y. (2024) · IEEE International Conference on Cognitive Machine Intelligence (CogMI) · doi:10.1109/cogmi62246.2024.00044 · Cited by 2
AI-Powered Data Interaction: A Natural Language Chatbot for CSV, Excel, and SQL Files
Dhanya, K. R., & Venkatesh, S. B. (2025) · Journal of Information Technology and Digital World · doi:10.36548/jitdw.2025.1.001 · Cited by 2
Generating creativity through ChatGPT: an empirical investigation in open innovation platforms
Li, Y., Xinyue, L., & Wang, L. (2025) · Information Technology and Management · doi:10.1007/s10799-025-00454-5 · Cited by 2
Reassessing AI in Medicine: Exploring the Capabilities of AI in Academic Abstract Synthesis
Wang, Z., & Zhou, C. (2024) · Journal of Medical Internet Research · doi:10.2196/55920 · Cited by 2
“However, there are concerns that ChatGPT 3.5 might not be ideally suited for processing lengthy texts, potentially leading to misunderstandings or the provision of inaccurate information [2].”
InfluentialSchnepper, R., Roemmel, N., Schaefert, R., Lambrecht-Walzinger, L., & Meinlschmidt, G. (2024) · doi:10.2196/preprints.57986 · Cited by 2
Ethical Challenges and Guidelines for AI Deployment in Healthcare
Kanetkar, S. R., & Thorat, V. M. (2024) · doi:10.1002/9781394272266.ch21 · Cited by 2
Palacios, D., & Rivadeneira, L. (2025) · Communications in Computer and Information Science · doi:10.1007/978-3-031-83210-9_21 · Cited by 1
Leveraging Artificial Intelligence in Scholarly Publishing
Fatima Zahra, O., Khouloud EL, M., & Soufiane, O. (2025) · Journal of Applied Social Sciences and Humanities · doi:10.71426/jassh.v1.i1.pp1-8 · Cited by 2
Hallucinations in large language models within academic contexts: A systematic review.
Lai, K. E., Mustaffa, N., Preece, C. N., & Wong, F. Y. (2026) · Journal of Science and Technology Policy Management · doi:10.1108/jstpm-01-2026-0015 · Cited by 1
Taxonomy of Mathematical Plagiarism
Satpute, A., Greiner-Petter, A., Gießing, N., Beckenbach, I., Schubotz, M., Teschke, O., Aizawa, A., & Gipp, B. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-56066-8_2 · Cited by 1
“Large Language Models (LLMs) make plagiarism easier due to produced unreferenced content [19].”
BackgroundChatGPT’s performance evaluation for annotating multi-label text in Indonesian language
Hakim, M. F. A., & Prasetiyo, B. (2024) · AIP conference proceedings · doi:10.1063/5.0210320 · Cited by 1
Empowering knowledge through AI: open scholarship proactively supporting well trained generative AI
Hellen, E. (2024) · Insights the UKSG journal · doi:10.1629/uksg.649 · Cited by 4
Aras, N., & Çalışkan, N. (2024) · International Journal of Urological Nursing · doi:10.1111/ijun.12428 · Cited by 1
EFFECTS OF USING AI IN TEACHING ENGLISH: A WORKPLACE-ENGLISH CASE STUDY
Kavgić, A. (2024) · ГОДИШЊАК ФИЛОЗОФСКОГ ФАКУЛТЕТА У НОВОМ САДУ · doi:10.19090/gff.v49i2.2534 · Cited by 2
Afsal, C. P., & Kuppusamy, K. S. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-97-5703-9_34 · Cited by 1
Conversations with a Purpose: The Impact of ChatGPT Among Higher Education Students
Planells-Artigot, E., Moll-López, S., & Nuñez-Pérez, A. (2024) · Springer eBooks · doi:10.1007/978-3-031-75140-0_29 · Cited by 1
ChatGPT and trustworthiness of artificial intelligence in medical education
Prashanth, G. P. (2025) · Journal of Education and Health Promotion · doi:10.4103/jehp.jehp_334_24 · Cited by 1
Kim, M., Kim, T., Chung, J., Choi, H., Min, S., Lim, J., & Park, S. (2025) · IEEE/ACM International Conference on Software Engineering: Software Engineering in Society (ICSE-SEIS) · doi:10.1109/icse-seis66351.2025.00022 · Cited by 1
“These studies collectively underscore the promising role that LLMs could play in enhancing educational practices and outcomes [15].”
Jiang, X., He, J., Wang, Y., Hao, S., Fang, W., & Li, T. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-6954-7_29 · Cited by 1
Veatch, O. J., Mathew, J., Rockowitz, S., Baldridge, D., Wetzel, A. S., Niarchou, M., Clarke, M., Shankar, P. R. V., Shankar, S. P., Cohen, J. S., et al. (2025) · Therapeutic Advances in Rare Disease · doi:10.1177/26330040251356521 · Cited by 1
“…from clinical notes with minimal training and development time (see Meyer et al., 2023 for descriptions of strengths and biases associated with LLMs 32 ), can be further fine-tuned with domain specific data, providing wide application in genetics and genomics across medical specialties (e.g.,…”
InfluentialSankar, B., & Sen, D. (2025) · Journal of Engineering Design · doi:10.1080/09544828.2025.2537465 · Cited by 1
“…language processing (NLP) have recently demonstrated their ability to generate rational, contextually relevant text responses based on their extensive training data, mimicking human-like reasoning (Chen et al. 2024; Jin et al. 2024; Meyer, Urbanowicz, and Martin 2023; Yuhan, Xiuying, and Rui2023 ).”
Trapped in the Prompt Loop: Reprompt Behavior in Writing with ChatGPT
Jung, H., Jeong, Y., & Lee, J. (2025) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2025.2530100 · Cited by 1
“ChatGPT and other LLMs operate based on publicly available internet text, which can lead to biased outputs reflecting stereotypes related to gender, race, and other factors depending on the training data (Meyer et al., 2023).”
Bashar, M. A., & Nayak, R. (2025) · Knowledge and Information Systems · doi:10.1007/s10115-025-02551-x · Cited by 1
Uddın, M. M. (2025) · Discover Education · doi:10.1007/s44217-025-00805-y · Cited by 3
“Meyer et al. [28] and Mrabet and Studholme [10] similarly observe that AI illiteracy—users’ lack of understanding about how AI tools function and what constitutes ethical usage—contrib-utes to unintentional plagiarism.”
Pérez-Portabella, A., Arias-Oliva, M., Sánchez, J. D. A., & Castillo, G. P. (2025) · Computers · doi:10.3390/computers14110458 · Cited by 4
Lessons from an AI-Sprint: a proposal for measuring human-AI cooperation in research
Wendering, L., Ratkovic, M., Gschwend, T., Schoenenberger, H., & Scaplehorn, N. (2025) · Humanities and Social Sciences Communications · doi:10.1057/s41599-025-06110-1 · Cited by 1
Adversarial Attack on Large Language Models using Exponentiated Gradient Descent
Biswas, S., Nishino, M., Chacko, S. J., & Liu, X. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227336 · Cited by 1
Chaves, L., Gonzaga, D., Tiago, L., Silva, A. P., & de Oliveira, F. C. M. (2025) · Simpósio Brasileiro de Qualidade de Software (SBQS) · doi:10.5753/sbqs.2025.13599 · Cited by 1
Addressing Challenges and Seizing Opportunities with ChatGPT in Computer Programming Education
Tang, C. M., Ng, V. S. C., Leung, H. M. F., & Yuen, J. C. H. (2025) · Signals and communication technology · doi:10.1007/978-3-031-96855-6_11 · Cited by 1
Zhou, A.-L., Chiang, J. Y.-H., Chan, K. S., Tan, N., & Shelat, V. G. (2025) · World Journal of Gastroenterology · doi:10.3748/wjg.v31.i46.111669 · Cited by 1
Hallucinations in Scholarly LLMs
Lamba, N., Tiwari, S., & Gaur, M. (2025) · Open Conference Proceedings · doi:10.52825/ocp.v8i.3175 · Cited by 3
Using large language model to aid in teaching medical imaging report writing
Chen, Y., Xiang, P., Zhou, Q., Li, C., Zhang, X., Wang, J., Wang, H. J., Gao, Z., Yang, Z., Ye, S., et al. (2025) · Medical Teacher · doi:10.1080/0142159x.2025.2603353 · Cited by 2
“Since medical diagnostic reports are highly textual, LLMs are particularly well suited to analyzing them [11,12].”
Mushtaq, S., & Veningston, K. (2026) · International Journal of Data Science and Analytics · doi:10.1007/s41060-025-00956-z · Cited by 2
Chapman, O., & Dennis, D. (2026) · Journal of Business & Finance Librarianship · doi:10.1080/08963568.2026.2614884 · Cited by 1
“Meyer et al. (2023) further highlights that while large language models such as ChatGPT expand opportunities for academic research and writing, they also raise concerns regarding bias, factual reliability, and overreliance on automation.”
A Comparative Study on Large Language Models with an Overview of Their Issues and Challenges
Kumar, A., & Sood, M. (2026) · Lecture notes in electrical engineering · doi:10.1007/978-981-96-5969-2_13 · Cited by 1
Generating Experimental Text Stimuli for Psychological Research Using ChatGPT
Lechuga, J., & Karle, N. N. (2026) · Advances in Methods and Practices in Psychological Science · doi:10.1177/25152459251407631 · Cited by 1
“Therefore, Chat-GPT’s generated text is arguably original and does not infringe on anyone’s intellectual property (Meyer et al., 2023).”
Teachers’ perceptions of generative AI in gender-inclusive STEM education: a grounded theory study
Tipple, C., Keane, T., Wang, T., & Cerovac, M. (2026) · Education and Information Technologies · doi:10.1007/s10639-026-13908-2 · Cited by 2
“Indeed, literature indicates that AI-generated content commonly produces outputs that reinforce gender-based stereotypes about STEM and stereotypical gender roles, evident in both textual and visual content (Kotek et al., 2023; Locke & Hodgdon, 2024; Meyer et al., 2023).”
Towards AI Literacy in L1 Education: Examining L1 Teachers’ Beliefs About Generative AI
Stevens, E., van Rijt, J., Bluijs, S., & Bax, S. (2026) · Digital studies in language and literature · doi:10.1515/dsll-2025-0003 · Cited by 2
“…acknowledge that students ’ use of GenAI might a ff ect the acquisition of literacy skills, such as basic reading and writing skills in primary and secondary education (Kasneci et al. 2023; Trust et al. 2023) and academic writing skills in higher education (Grassini 2023; Meyer et al. 2023).”
Metadata
- 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
- Social Sciences › Education › Artificial Intelligence in Education
- Physical Sciences › Computer Science › Artificial Intelligence › 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.
A language model drew the claims from the paper on 2026-10-09, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-09 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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
