CoChat
Sign Up
PreprintCited by 238Artificial Intelligence in Healthcare and EducationArtificial Intelligence in EducationArtificial Intelligence Applications

An exploratory survey about using ChatGPT in education, healthcare, and research

Open access
Hosseini, Mohammad, Gao, Catherine A., Liebovitz, David M., Carvalho, Alexandre M., Ahmad, Faraz S., Luo, Yuan, MacDonald, Ngan, Holmes, Kristi L. +1 more·2023·PLoS ONE
DOI 10.1101/2023.03.31.23287979PMC PMC10104227
Open this paper in CoChat — free accountOpen PDF

Abstract

Objective ChatGPT is the first large language model (LLM) to reach a large, mainstream audience. Its rapid adoption and exploration by the population at large has sparked a wide range of discussions regarding its acceptable and optimal integration in different areas. In a hybrid (virtual and in-person) panel discussion event, we examined various perspectives regarding the use of ChatGPT in education, research, and healthcare. Materials and Methods We surveyed in-person and online attendees using an audience interaction platform (Slido). We quantitatively analyzed received responses on questions about the use of ChatGPT in various contexts. We compared pairwise categorical groups with Fisher's Exact. Furthermore, we used qualitative methods to analyze and code discussions. Results We received 420 responses from an estimated 844 participants (response rate 49.7%). Only 40% of the audience had tried ChatGPT. More trainees had tried ChatGPT compared with faculty. Those who had used ChatGPT were more interested in using it in a wider range of contexts going forwards. Of the three discussed contexts, the greatest uncertainty was shown about using ChatGPT in education. Pros and cons were raised during discussion for the use of this technology in education, research, and healthcare. Discussion There was a range of perspectives around the uses of ChatGPT in education, research, and healthcare, with still much uncertainty around its acceptability and optimal uses. There were different perspectives from respondents of different roles (trainee vs faculty vs staff). More discussion is needed to explore perceptions around the use of LLMs such as ChatGPT in vital sectors such as education, healthcare and research. Given involved risks and unforeseen challenges, taking a thoughtful and measured approach in adoption would reduce the likelihood of harm.

AI snapshot

Written from the paper’s full text.

TL;DR

In a hybrid event survey, 420 respondents (estimated response rate 49.7%) reported limited experience with ChatGPT, with only 40% having tried it and trainees more likely than faculty to have used it. Prior users were more interested in and accepting of ChatGPT across contexts, while uncertainty was greatest for education.

Key findings

  • Only 40% of the audience had tried ChatGPT. Medical trainees were more likely to have tried it (64.2%) than clinical faculty (31.1%), p=0.03; graduate students/postdocs (56.6%) versus research faculty (49.2%) was not significant, p=0.46.
  • Those who had used ChatGPT had higher interest in using it going forward: 39.9% were interested “to a great extent” versus 15.9% of non-users, p<0.001.
  • Prior users were more likely to deem ChatGPT acceptable for research (89.3% vs 75%, p<0.001), healthcare (62.5% vs 48.8%, p=0.008), and education (63.9% vs 30.2%, p<0.001).
  • Uncertainty about use was greater for education and healthcare than research: “I don’t know, it is too early to make a statement” was selected by 226 (53.8%) for education, 177 (42.2%) for healthcare, and 75 (17.9%) for research (p<0.001 for both comparisons with research). For healthcare, 42% approved of administrative uses and 12.2% thought it could be used for any purpose.

Methods

  • Hybrid (virtual and in-person) panel event survey using Slido; 420 responses were collected, with a conservative estimated response rate of 49.7%. The Northwestern IRB granted exemption, and data were anonymized.
  • Quantitative analysis was performed in Python v3.8 with scipy, matplotlib, seaborn, tableone, and plot_likert. Responses were binarized as “yes” versus “No + unsure,” and pairwise categorical groups were compared with Fisher’s Exact tests. For 18 respondents with multiple roles, the most senior and most clinical role was used.
  • Qualitative analysis transcribed the session and inductively coded transcripts using three topic areas (education, healthcare, research), then three subcodes per area (possible positive impacts, possible negative impacts, remaining questions), for nine codes total; panel feedback was sought on the draft report.

Limitations

  • Stated in the paper: the large “Other” respondent category could not be broken down further; respondents were likely interested in ChatGPT because they registered for and attended the event, so results may not represent the academic community; medical trainees were the smallest group (3.3% of respondents), and their positive views may reflect role-related interest.
  • Evident from the text: this was a single-event, self-selected convenience sample, limiting generalizability; the cross-sectional survey and discussion cannot establish causal relationships or track changes over time.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1101/2023.03.31.23287979

Cites

20 of 20 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 10 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    There was greater uncertainty about using ChatGPT in Education and Healthcare compared to Research.

    “In response to questions about whether ChatGPT can be used, there was greater uncertainty around its use in Healthcare and Education, compared to using it in Research.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Those with prior experience thought ChatGPT was acceptable to use in healthcare (62.5% vs 48.8%, 13.7% higher, p=0.008).

    “Similarly, those with prior experience thought it was acceptable to use in healthcare 62.5% vs 48.8%, 13.7% higher, p=0.008.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Those who had already used ChatGPT were more likely to deem it acceptable for research purposes (89.3% vs 75%, 14.3% higher, p<0.001).

    “Those who had already used ChatGPT were more likely to deem it acceptable for research purposes (89.3% 'yes') versus those who had not used it before (75% ‘yes’), 14.3% higher, p<0.001 (Figure 6).”
    Quote in the paperNot a substantial claim
  • Findingown data

    Research trainees (graduate students and postdoctoral researchers) had a similar rate of having tried ChatGPT compared with research faculty (56.6% vs 49.2%, p=0.46).

    “A more similar number of graduate students and postdoctoral researchers had tried ChatGPT (56.6%) compared with research faculty (49.2%), p=0.46.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Across all roles except medical trainees, the most common response regarding interest in using ChatGPT going forwards was 'Somewhat'.

    “Across all roles except medical trainees, the most common response regarding interest in using ChatGPT going forwards was ‘Somewhat’ (Figure 3).”
    Quote in the paperNot a substantial claim
  • May combine two claims. 17
  • Classified as probably not a substantial claim. 19

  • May combine two claims.

    Findingown data

    For Research, 17.9% (75) selected 'I don't know, it is too early to make a statement', compared with 53.8% (226) for Education (p<0.001) and 42.2% (177) for Healthcare (p<0.001).

    “For Research, only 75 (17.9%) of respondents selected ‘I don’t know, it is too early to make a statement’, compared with 226 (53.8%) when asked about using it in Education (p<0.001), and 177 (42.2%), when asked about using it in Healthcare (p<0.001), (Table 1, Supplemental Document).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Those who had used ChatGPT had higher interest in using it going forwards compared with those who had not; 39.9% had interest 'to a great extent' vs 15.9%, p<0.001.

    “Those who had used ChatGPT already had higher interest in using it compared with those who did not; 39.9% had interest in using it ‘to a great extent’ compared with 15.9%, p<0.001) (Figure 4).”
    • Also stated in: § Results · plain
      “Those who had used ChatGPT were more interested in using it in a wider range of contexts going forwards.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Medical and research trainees were more interested in using ChatGPT for education purposes compared with clinical and research faculty, though this was not statistically significant.

    “Medical and research trainees were more interested in using it for education purposes compared with clinical and research faculty, though this was not statistically significant.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The largest group of respondents identified as 'Other' (173, 41.2%).

    “The largest group of respondents identified as ‘Other’, with 173 respondents (41.2% of all respondents).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Pros and cons were raised during discussion for the use of ChatGPT in education, research, and healthcare.

    “Pros and cons were raised during discussion for the use of this technology in education, research, and healthcare.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    When responding about using ChatGPT in Healthcare, 42% of respondents approved of using it for administrative purposes, and 12.2% thought it could be used for any purpose.

    “Of note, when responding to the question about using ChatGPT in Healthcare, a significant portion (42% of respondents) of respondents approved of using it for administrative purposes (for example, writing letters to insurance companies) and there was a smaller population of respondents who thought it could be used for any purpose (12.2%) (Table 1, Supplemental Document).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Medical trainees were the smallest group of respondents (14, 3.3%).

    “The smallest group were medical trainees (medical students, residents, and fellows) at 14 respondents (3.3% of all respondents)”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    The broad interest and engagement sparked by ChatGPT strongly suggests that, while a work in progress, LLMs have a significant potential for disruption.

    “The broad interest and engagement sparked by ChatGPT strongly suggests that, while a work in progress, LLMs have a significant potential for disruption.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: ChatGPT is the first large language model to reach a large, mainstream audience.
    • Rests on an assumption: ChatGPT's rapid adoption and exploration by the population at large has sparked a wide range of discussions regarding its acceptable and optimal integration in different areas.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    A limitation of the study is the inability to break down and better delineate the large 'Other' category of respondents.

    “Some of the limitations include inability to break down and better delineate the large “Other” category of respondents.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    There was a range of perspectives around the uses of ChatGPT in education, research, and healthcare, with still much uncertainty around its acceptability and optimal uses.

    “There was a range of perspectives around the uses of ChatGPT in education, research, and healthcare, with still much uncertainty around its acceptability and optimal uses.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Trainees were more interested in using ChatGPT than faculty, having more positive views, interest, and acceptability beliefs in using the technology.

    “Trainees were more interested in using ChatGPT than faculty, having more positive views, interest, and acceptability beliefs in using the technology.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Although medical trainees had positive views towards ChatGPT and its use, they were the smallest group of respondents (3.3% of the cohort).

    “Although medical trainees had positive views towards ChatGPT and its use, they were our smallest group of respondents (3.3% of our cohort).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The Northwestern IRB granted exemption (STU00218786).

    “The Northwestern IRB granted exemption (STU00218786).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    The event had 1,174 registrations, a peak of 718 webinar participants, and 126 people indicating they would attend in-person.

    “We had 1,174 people register for the event. The peak number of webinar participants during the event was 718, and 126 people indicated they would attend in-person.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Anonymized data were collected using a paid version of Slido.

    “We collected anonymized data using a paid version of Slido (Bratislava, Slovakia; https://www.sli.do/).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The survey received 420 responses from an estimated 844 participants, a response rate of 49.7%.

    “We received 420 responses from an estimated 844 participants (response rate 49.7%).”
    • Also stated in: § Survey results · hedged
      “We received survey responses from 420 people; a conservative estimated response rate is 49.7%.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Participants who had used the technology before had a tendency to have a more optimistic outlook about LLMs in general, whereas never-users seemed to have more concerns about its widespread adoption.

    “Participants that had used the technology before had a tendency to have a more optimistic outlook about LLMs in general whereas never-users seemed to have more concerns about its widespread adoption.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Participants were informed about survey details and offered a chance to view the information sheet and consent form before the survey.

    “Participants were informed about the survey details, such as anonymized data collection and voluntary participation, and were offered a chance to view the information sheet and consent form before the start of the survey.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The findings point to a potential generational divide between early adopters (trainees) and late adopters (faculty), with the latter in positions of power to dictate policy to trainees and the academic community at large.

    “This points to a potential generational divide between early adopters (trainees) and late adopters (faculty), with the latter in positions of power to dictate policy to trainees and the academic community at large.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

    Interpretationreasoning

    Detection applications such as the OpenAI Classifier seem unreliable and for the foreseeable future will likely remain so.

    “Detection applications such as the OpenAI Classifier, which uses four (ambiguous) categories to label inputted text (Very unlikely, Unlikely, Unclear if it is/Possibly or Likely AI-generated) seem unreliable and for the foreseeable future will likely remain so.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    The recruitment material for the event took a neutral tone to the technology, as evidenced by respondents from other roles who had more lukewarm or uncertain feelings towards ChatGPT.

    “We took a neutral tone to the technology in our recruitment material for the event, as evidenced by the respondents from other roles who had more lukewarm or uncertain feelings towards ChatGPT.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The authors suspect there is high interest from medical trainees inherent to their role.

    “Hence, we suspect there is high interest from this group inherent to their role.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Even before the introduction of OpenAI's ChatGPT, using computer generated text in academic publications had an estimated prevalence of 4.29 papers for every one million papers as per 2021.

    “Even before the introduction of OpenAI’s ChatGPT, using computer generated text in academic publications had an estimated prevalence of “4.29 papers for every one million papers” as per 2021[10].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [10] Cabanac G, Labbé C, Magazinov A. Tortured phrases: A dubious writing style emerging in science. Evidence of critical issues affecting established journals. 2021. doi: 10.48550/arXiv.2107.06751

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    As ChatGPT and other recently developed applications based on LLMs mainstream the use of AI-generated content, detection will likely become much more difficult.

    “As ChatGPT and other recently developed applications based on LLMs mainstream the use of AI-generated content, detection will likely become much more difficult.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    There were concerns about the negative impact of using LLMs on the integrity of academic publications.

    “There were concerns about the negative impact of using LLMs on the integrity of academic publications[11].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [11] Cabanac G, Labbé C. Prevalence of nonsensical algorithmically generated papers in the scientific literature. J Assoc Inf Sci Technol 2021; 72: 1461– 76. doi: 10.1002/asi.24495

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Some New York schools banned students from using ChatGPT, while others adopted policies in their syllabus that encourage students to engage with these models as long as they disclose it.

    “For instance, some New York schools banned students from using ChatGPT[1], while others adopted policies in their syllabus that encourage students to engage with these models as long as they disclose it[2].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [1] Yang M. New York City schools ban AI chatbot that writes essays and answers prompts. The Guardian. 2023. https://www.theguardian.com/us-news/2023/jan/06/new-york-city-schools-ban-ai-chatbot-chatgpt (accessed 27 Mar 2023).

      Not checked in the free run.

    • cites [2] Mollick E. All my classes suddenly became AI classes. One Useful Thing Also Some Things. 2023. https://oneusefulthing.substack.com/p/all-my-classes-suddenly-became-ai (accessed 25 Jan 2023).

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Various journal editors and professional societies have developed disclosure guidelines, stressing that LLMs cannot be authors, and suggesting that disclosure should happen as part of the methods section, describing who used the system, when, and using which prompts plus adding it among cited references.

    “Various journal editors and professional societies have developed disclosure guidelines, stressing that LLMs cannot be authors[13,14], and suggesting that disclosure should happen as part of the methods section, describing who used the system, when, and using which prompts plus adding it among cited references[15].”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [13] Flanagin A, Bibbins-Domingo K, Berkwits M, Nonhuman “Authors” and Implications for the Integrity of Scientific Publication and Medical Knowledge. JAMA 2023; 329: 637– 9. doi: 10.1001/jama.2023.1344 36719674

      Not checked in the free run.

    • cites [14] Zielinski C, Winker M, Aggarwal R, Chatbots, ChatGPT, and Scholarly Manuscripts: WAME Recommendations on ChatGPT and Chatbots in Relation to Scholarly Publications. 2023. https://wame.org/page3.php?id=106 37615142

      Not checked in the free run.

    • cites [15] Hosseini M, Rasmussen LM, Resnik DB. Using AI to write scholarly publications. Account Res 2023; 0: 1– 9. doi: 10.1080/08989621.2023.2168535 PMC10366336 36697395

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Some educators fed ChatGPT questions from a freely available United States Medical Licensing Examination (USMLE) and reported a near or at passing range performance.

    “Some educators fed ChatGPT questions from a freely available United States Medical Licensing Examination (USMLE) and reported a near or at passing range performance[3].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [3] Kung TH, Cheatham M, Medenilla A, Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLOS Digit Health 2023; 2: e0000198. doi: 10.1371/journal.pdig.0000198 36812645 PMC9931230

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    LLMs can be used in scholarly reviews to support editorial practices, but various risks such as inaccuracies and biases require researchers to engage with LLMs cautiously.

    “Besides writing scholarly manuscripts, LLMs can also be used in scholarly reviews to support editorial practices,e.g., supporting the search for suitable reviewers, the initial screening of manuscripts, and the write-up of final decision letters from individual review reports, but various risks such as inaccuracies and biases require researchers to engage with LLMs cautiously[16].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [16] Hosseini M, Horbach SPJM. Fighting reviewer fatigue or amplifying bias? Considerations and recommendations for use of ChatGPT and other Large Language Models in scholarly peer review. 2023. doi: 10.21203/rs.3.rs-2587766/v1 PMC10191680 37198671

      Not checked in the free run.

29 of 29

  • The paper relies on this without support.

    ChatGPT is the first large language model to reach a large, mainstream audience.

    “ChatGPT is the first large language model (LLM) to reach a large, mainstream audience.”
    “The introduction of OpenAI's ChatGPT has delivered large language model (LLM) systems to a mainstream audience.”

  • The paper relies on this without support.

    ChatGPT's rapid adoption and exploration by the population at large has sparked a wide range of discussions regarding its acceptable and optimal integration in different areas.

    “Its rapid adoption and exploration by the population at large has sparked a wide range of discussions regarding its acceptable and optimal integration in different areas.”
2 of 2

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?

  • Although the number of respondents exceeded the required sample size, the response rate was lower compared with similar studies, which may limit the generalizability of the findings.

    “First, although the number of respondents exceeded the required sample size, the response rate was lower compared with similar studies [19, 21, 22, 24, 25, 29, 31, 34, 37, 40, 62], which may limit the generalizability of the findings.”

    From Knowledge and use, Perceptions of Benefits and Limitations of Artificial Intelligence Chatbots among Italian Physiotherapy students: a cross-sectional National Study (2025) · cites this paper as [21]

    The Survey results section reports a 49.7% response rate, providing a figure the citing authors could use for comparison. The source does not establish that the citing study’s response rate was lower or that its respondent count exceeded a required sample size. Its Limitations section raises a representativeness concern about its own survey, but attributes that concern to likely participant interest in ChatGPT, not to the response rate. (LLM, conf=0.83)

    Experimental — accuracy not yet measured

  • Italian physiotherapy students are still in an exploratory phase with AI Chatbots, showing a prevailing sense of uncertainty and skepticism and having yet to adopt a clear stance on the technology's advantages and disadvantages.

    “As reported in the literature [21, 43], these results highlight a prevailing sense of uncertainty and skepticism, suggesting that Italian students are still in an exploratory phase with this emerging technology and have yet to adopt a clear stance on its advantages and disadvantages.”

    From Knowledge and use, Perceptions of Benefits and Limitations of Artificial Intelligence Chatbots among Italian Physiotherapy students: a cross-sectional National Study (2025) · cites this paper as [21]

    The Discussion reports uncertainty about ChatGPT’s acceptability, and the Results describe respondents who felt it was too early to judge its use in education. These passages support the comparison to uncertainty in the literature, but this survey did not study Italian physiotherapy students; the conclusion about those students comes from the citing authors’ own results. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • There is a notable scarcity of literature that evaluates the proficiency of Google's Bard and ChatGPT specifically in the field of endocrinology, diabetes, and diabetes technology.

    “Nevertheless, there is a notable scarcity of literature that evaluates the proficiency of Google's Bard and ChatGPT specifically in the field of endocrinology, diabetes, and diabetes technology.7,8”

    From The Scientific Knowledge of Bard and ChatGPT in Endocrinology, Diabetes, and Diabetes Technology: Multiple-Choice Questions Examination-Based Performance (2025) · cites this paper as [8]

    The Introduction’s healthcare section discusses general medical uses of ChatGPT, including answering questions and generating clinical vignettes. Our copy does not assess the literature on Bard and ChatGPT proficiency in endocrinology, diabetes, or diabetes technology; searches for those topics found no relevant passage. No page numbers were provided. (LLM, conf=0.76)

    Experimental — accuracy not yet measured

  • LLMs are not able to replicate a process of clinical reasoning in general and fail to comprehend the complexity of the process.

    “This is in line with others noting that LLMs are not able to replicate a process of clinical reasoning in general and, hence, fail to comprehend the complexity of the process44,59,75.”

    From The ethics of ChatGPT in medicine and healthcare: a systematic review on large language models (LLMs) (2024) · cites this paper as [44]

  • Power concentration means pricing control lies with LLM companies, with revenues predominantly directed towards them.

    “Power concentration also means pricing control lies with LLM companies, with revenues predominantly directed towards them44.”

    From The ethics of ChatGPT in medicine and healthcare: a systematic review on large language models (LLMs) (2024) · cites this paper as [44]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Hosseini, M., Gao, C. A., Liebovitz, D. M., Carvalho, A. M., Ahmad, F. S., Luo, Y., MacDonald, N., Holmes, K. L., & Kho, A.
Affiliations
Northwestern University (US)
Corresponding author
Hosseini, Mohammad
ORCID
Venue
PLoS ONE
Publisher
openRxiv
Published
2023-04-03
Volume
18
Issue
10
Pages
e0292216-e0292216
ISSN
1932-6203 (ISSN-L)
DOI
10.1101/2023.03.31.23287979
PMC
PMC10104227
OpenAlex
W4387363029
Cited by
238
References
20
Citations by year
2026: 38 · 2025: 76 · 2024: 95 · 2023: 29
FWCI
9.48
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • medRxiv (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • American Heart Association (856917, AHA number 856917)
  • Northwestern University (UL1 TR001422)
  • National Institutes of Health (UL1TR001422)
  • National Heart, Lung, and Blood Institute (K23HL155970, 856917, F32HL162377)
  • National Institute of Diabetes and Digestive and Kidney Diseases (P30DK092949)
  • National Center for Advancing Translational Sciences
  • NHLBI Division of Intramural Research (F32HL162377)
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence Applications
Keywords
ChatGPT, large language models, generative AI adoption, artificial intelligence in education, artificial intelligence in healthcare, artificial intelligence in research, user perception, technology acceptance, AI risks

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