Patient perspectives on informed consent for medical AI: A web-based experiment
Open accessThe DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index; 10 of the 37 sources its claims cite were checked for support.
Abstract
Objective: Despite the increasing use of AI applications as a clinical decision support tool in healthcare, patients are often unaware of their use in the physician's decision-making process. This study aims to determine whether doctors should disclose the use of AI tools in diagnosis and what kind of information should be provided. Methods: A survey experiment with 1000 respondents in South Korea was conducted to estimate the patients’ perceived importance of information regarding the use of an AI tool in diagnosis in deciding whether to receive the treatment. Results: The study found that the use of an AI tool increases the perceived importance of information related to its use, compared with when a physician consults with a human radiologist. Information regarding the AI tool when AI is used was perceived by participants either as more important than or similar to the regularly disclosed information regarding short-term effects when AI is not used. Further analysis revealed that gender, age, and income have a statistically significant effect on the perceived importance of every piece of AI information. Conclusions: This study supports the disclosure of AI use in diagnosis during the informed consent process. However, the disclosure should be tailored to the individual patient's needs, as patient preferences for information regarding AI use vary across gender, age and income levels. It is recommended that ethical guidelines be developed for informed consent when using AI in diagnoses that go beyond mere legal requirements.
Overview
CoChat Summary
TL;DR
In a web-based experiment with 1000 South Korean respondents, using an AI tool in diagnosis increased the perceived importance of information compared with referral to a human radiologist. The study supports disclosing AI use during informed consent, with disclosure tailored to patient characteristics such as gender, age, and income.
Key findings
- AI use significantly increased perceived importance of most information items versus radiologist scenarios, including “whether an AI tool was used” (mean difference = −0.2863, p < .001), “doctor experience” (−0.2832, p < .001), “generally used” (−0.2634, p < .001), and “opt out” (−0.3335, p < .001); H1 was supported except for long-term effects and whether opinions were the same.
- AI information when AI was used was rated more important than or similar to the lowest-rated regularly disclosed surgery information (“short-term effect”): performance (diff = −0.1895, p = .01), doctor experience (−0.2098, p = .005), and opinion same (−0.2443, p = .001).
- High risk of the recommended treatment increased importance only for “opinion same” (b = 0.0488, p = .008) and “reason reject/accept” (b = 0.0434, p = .02); prevalence of AI use and superior AI performance had no significant effect, so H3 and H4 were not supported.
- Gender, age, and income had statistically significant effects on every piece of AI information: female β = 0.1888–0.2695, age β = 0.0985–0.1540, income β = 0.0723–0.1131 (all p < .001 or p < .01 depending on item).
Methods
- Survey experiment conducted online in June 2022 with 1000 respondents aged 20–69 from the Embrain national research panel in South Korea; quotas were set at 100 per gender × five age groups.
- Participants were randomly assigned five vignettes each (one radiologist scenario and four AI scenarios) from 10 total scenarios varying AI vs radiologist, high vs low risk, common vs rare use, and superior vs average performance.
- Outcomes were 15 information items rated on a 7-point Likert scale; analysis used multilevel mixed-effect linear regression and linear regression with robust standard errors clustered in respondents.
Limitations
- The paper states the commercial panel may introduce selection bias toward technology affinity, and the sample overrepresented undergraduate education (66.8% vs 50.5% in the 2020 census) and graduate education (11.2% vs 6.1%).
- The paper states it measured self-reported familiarity with AI, not actual knowledge; most information was rated very or extremely important, so statistically significant differences may not be clinically relevant; it did not directly test whether providing all desired AI information affects decisions to undergo surgery; and South Korean results may not generalize.
- Evident from the design: hypothetical vignettes and self-reported importance ratings capture stated preferences rather than observed informed-consent behavior.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
6 Valid · 1 Not valid · 9 Not checked · 63 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“At present, there is no ethical and legal consensus regarding whether disclosing the application of a medical AI is required for informed consent in the United States, the European Union, or South Korea.”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Introduction > Background
What the sources say
- Partly supportedcites [7] Amann J Blasimme A Vayena E, et al. Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC Med Inform Decis Mak 2020; 20: 310. 33256715 PMC7706019 sourceExperimental — accuracy not yet measured
The Ethical implications section states the ethical half almost verbatim: 'At the moment, an ethical consensus has not yet emerged as to whether disclosing the use of an opaque medical AI algorithm should be a mandatory requirement of informed consent' — though unqualified by jurisdiction and limited to *opaque* algorithms, whereas the claim generalises to 'the application of a medical AI'. The legal half is covered only for the US ('For the US, as Cohen laid out, there is currently no clear-cut answer to what extent the integration of ML/DL into clinical decision-making has to be disclosed' — The legal perspective, and there attributed to Cohen's analysis rather than original to this paper) and for the EU indirectly (a 'current debate' over whether GDPR requires explainable AI, with 'currently ambiguous phrasings'; FDA/MDR 'rather vaguely requiring explainability'). The paper explicitly frames its legal analysis 'From a Western legal point-of-view', and South Korea appears nowhere in our copy (searched: South Korea, Korea/Asian regulation, legal consensus disclosure), so the Korean limb of the claim is not supported by this source. (LLM, conf=0.82)
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.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Although the value of empirical research on patients’ views regarding information disclosure in the use of AI/ML in medicine has been recognized, 12 no research has yet explored what information patients consider important or significant for making treatment decisions when AI is used in a clinical setting.26–28”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Introduction > Prior work
What the sources say
- Partly supportedcites [26] Ongena YP Hommes S Kwee TC, et al. A qualitative study to understand patient perspective on the use of artificial intelligence in radiology. J Am Coll Radiol 2019; 16: 1416– 1419. 30878311 sourceExperimental — accuracy not yet measured
The ‘Description of the Problem’ says patients’ views of AI in radiology were still unknown, and ‘What Was Done’ notes a lack of specific literature on those views. The paper does not establish that no research in any clinical setting had explored which information patients consider important for treatment decisions; its own interviews concern radiology procedures and scan results. (LLM, conf=0.78)
- Could not verifycites [27] Nelson CA Pérez-Chada LM Creadore A, et al. Patient perspectives on the use of artificial intelligence for skin cancer screening: a qualitative study. JAMA Dermatol 2020; 156: 501– 512. 32159733 PMC7066525 sourceExperimental — accuracy not yet measured
The Abstract describes a study of patients’ views on AI for skin cancer screening, and the Discussion notes a paucity of data on those views. Neither passage establishes that no research had explored what information patients consider important for treatment decisions when AI is used clinically. Searches for information disclosure and treatment decisions found no passage making that claim. (LLM, conf=0.73)
- Partly supportedcites [28] Iqbal JD Christen M. The use of artificial intelligence applications in medicine and the standard required for healthcare provider-patient briefings-an exploratory study. Digit Health 2022; 8: 20552076221147423. 36601281 PMC9806394 sourceExperimental — accuracy not yet measured
The Introduction (‘Contribution of this study’; page unavailable) says empirical research on AI-related healthcare provider–patient briefings has been lacking. It does not specifically establish that no research has explored which information patients consider important for treatment decisions. The paper’s own survey asked medical students, physicians and bioethicists, not patients. (LLM, conf=0.83)
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.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Professional bodies have not offered a protocol for obtaining informed consent when using AI tools, which could potentially aid in developing an AI-specific standard of care.”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Introduction > Background
What the sources say
- Supportedcites [11] Harvey HB Gowda V. Clinical applications of AI in MSK imaging: a liability perspective. Skeletal Radiol 2022; 51: 235– 238. 33835241 source
The Informed consent section states that 'an AI-specific standard of care does not yet exist' and that its development 'could be further aided by professional bodies, such as the American College of Radiology (ACR) or Society of Skeletal Radiology (SSR), which can articulate a specialty-wide protocol for obtaining informed consent' — the prospective framing (they can articulate such a protocol) carries the claim's point that no such protocol has yet been offered. The same section notes that 'courts and state legislatures have yet to articulate an AI-specific standard for informed consent,' and the Conclusion repeats the forward-looking call for 'ACR and SSR-validated guidelines,' confirming the absence is the paper's premise rather than a report of an existing protocol. (LLM, conf=0.80)
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.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Courts and state legislatures have yet to articulate an AI-specific standard for informed consent.”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Introduction > Background
What the sources say
- Supportedcites [11] Harvey HB Gowda V. Clinical applications of AI in MSK imaging: a liability perspective. Skeletal Radiol 2022; 51: 235– 238. 33835241 source
The Informed consent section (p. of our copy, passage [4]) states verbatim that 'The scope of each is unclear; courts and state legislatures have yet to articulate an AI-specific standard for informed consent, a matter which remains contested in the literature.' Passage [3] reinforces this, noting that 'an AI-specific standard of care does not yet exist' and that professional bodies such as the ACR or SSR would need to articulate a specialty-wide consent protocol. (LLM, conf=0.95)
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.
- Supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“In the US, the Food and Drug Administration (FDA), which addresses the safety and effectiveness of ‘Software as a Medical Device (SaMD)’, does not address informed consent.8–10”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Introduction > Background
What the sources say
- Supportedcites [8] Ursin F Timmermann C Orzechowski M, et al. Diagnosing diabetic retinopathy with artificial intelligence: what information should be included to ensure ethical informed consent? Front. Med 2021; 8: 695217. PMC8333706 34368192 source
The Discussion section, under “Voluntariness and the Right to a Second Opinion,” explicitly says that the FDA in the USA addresses the safety and effectiveness of Software as Medical Device but does not address informed consent. This directly supports the attributed claim. (LLM, conf=0.94)
- Supportedcites [9] Martinez-Martin N Kreitmair K. Ethical issues for direct-to-consumer digital psychotherapy apps: addressing accountability, data protection, and consent. JMIR Ment Health 2018; 5: e32. PMC5938696 29685865 source
The Introduction (page unavailable) says some US digital therapy apps may be regulated by the FDA as medical devices and that increased FDA regulation would not address informed consent, despite the importance of safety and effectiveness. This supports the substance of the citing sentence. (LLM, conf=0.88)
- Not checked yetcites [10] Morley J Machado CCV Burr C, et al. The ethics of AI in health care: a mapping review. Soc Sci Med 2020; 260: 113172. 32702587 source
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.
- Supported
32 more claims
Create a free account to see all 79 claims and to verify the 27 references still open.
- Probably not a substantial claim. 27
- It may state two separate claims as one. 24
- The quote only partly states this claim. 1
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
“Some institutions implement AI technology as part of routine QP activities without obtaining explicit consent for each use, instead relying on broader policies and general consent from patients for treatment and quality assurance9.”
From Preventing unrestricted and unmonitored AI experimentation in healthcare through transparency and accountability (2025) · cites this paper as [9]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Currently, there is no universally agreed-upon ethical or legal requirement to disclose the use of medical AI for informed consent in the United States, the European Union, or South Korea8,9.”
From Preventing unrestricted and unmonitored AI experimentation in healthcare through transparency and accountability (2025) · cites this paper as [9]
The cited source says what the claim attributes to it.
In the citing paper
“Additionally, there is no established ethical or legal requirement in the United States for obtaining informed consent for the use of medical AI in clinical practice.”
From Preventing unrestricted and unmonitored AI experimentation in healthcare through transparency and accountability (2025) · cites this paper as [9]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Existing ethical regulations governing research do not apply to the routine deployment of AI to improve operational efficiency or to reduce the cognitive load of healthcare providers which are not considered health products despite risk of harm to patient outcomes, raising significant ethical concerns8,9.”
From Preventing unrestricted and unmonitored AI experimentation in healthcare through transparency and accountability (2025) · cites this paper as [9]
Experimental — accuracy not yet measured
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
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Associations of importance ratings with demographic features
“For example, respondents with higher education found architecture (β −0.211, P = 0.001) less important.”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Associations of importance ratings with demographic features
“| education | −0.0428 | .12 | −0.0935 | .001 | −0.0498 | .09 | −0.0206 | .47 | −0.0272 | .34 | −0.0183 | .52 | 0.0017 | .95 | −0.0364 | .23 |”
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
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Importance rating of information
“The two least important categories of information were the right to opt out from referring to a radiologist or AI decision support tool (mean score = 5.38) and architecture of the AI tool (mean score = 5.59).”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Importance rating of information
“| used | 5.5186 | 5.3770 | 5.6633 | −.2863 | <.001 |”
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
“Some 50 responses were collected until satisfying the quota of 6 individuals per ten (2 5) combinations of gender (female and male) and age group (20 s, 30 s, 40 s, 50 s and 60 s).”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Methods > Pilot study
Reported 50 · recomputed 60
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
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Comparison of importance of AI information and surgery information
“| short_term | 504 | 5.6766 | generally_used | 493 | 5.7972 | −.1206 | −.2687 | −.0275 | .11 |”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Comparison of importance of AI information and surgery information
“| Surgery information | obs | mean | AI information | obs | mean | diff | 95% CI min | 95% CI max | P value (\|T\| > \|t\|) |”
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
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Respondents’ characteristics
“| I visit hospital whenever I feel ill | Mean (SD) | | 1–5 scale | 2.4 (9.2) |”
§ Patient perspectives on informed consent for medical AI: A web-based experiment > Results > Respondents’ characteristics
“On a 1–5 rating scale (1 = Not at all, 5 = very much so), respondents chose 2.4 on average that they visit a hospital whenever they feel ill (risk_averse)”
3 more findings
Create a free account to see every finding of this check.
Citations
Cites
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362
Artificial intelligence in radiology
Hosny, A., Parmar, C., Quackenbush, J., Schwartz, L. H., & Aerts, H. J. W. L. (2018) · Nature Reviews Cancer · doi:10.1038/s41568-018-0016-5 · Cited by 3,803
Explainability for artificial intelligence in healthcare: a multidisciplinary perspective
the Precise4Q consortium, Amann, J., Blasimme, A., Vayena, E., Frey, D., & Madai, V. I. (2020) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-020-01332-6 · Cited by 2,203
Resistance to Medical Artificial Intelligence
Longoni, C., Bonezzi, A., & Morewedge, C. K. (2019) · Journal of Consumer Research · doi:10.1093/jcr/ucz013 · Cited by 2,015
The ethics of AI in health care: A mapping review
Morley, J., Machado, C. C. V., Burr, C., Cowls, J., Joshi, I., Taddeo, M., & Floridi, L. (2020) · Social Science & Medicine · doi:10.1016/j.socscimed.2020.113172 · Cited by 1,105
Nadarzynski, T., Miles, O., Cowie, A., & Ridge, D. (2019) · DIGITAL HEALTH · doi:10.1177/2055207619871808 · Cited by 810
Kiseleva, A., Kotzinos, D., & De Hert, P. (2022) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2022.879603 · Cited by 278
Nelson, C. A., Pérez-Chada, L. M., Creadore, A., Li, S. J., Lo, K., Manjaly, P., Pournamdari, A. B., Tkachenko, E., Barbieri, J. S., Ko, J. M., et al. (2020) · JAMA Dermatology · doi:10.1001/jamadermatol.2019.5014 · Cited by 278
Artificial intelligence in healthcare: a critical analysis of the legal and ethical implications
Schönberger, D. (2019) · International Journal of Law and Information Technology · doi:10.1093/ijlit/eaz004 · Cited by 230
Public Perception of Artificial Intelligence in Medical Care: Content Analysis of Social Media
Gao, S., He, L., Chen, Y., Li, D., & Lai, K. (2020) · Journal of Medical Internet Research · doi:10.2196/16649 · Cited by 221
Research works printed but not linked: 7
- Cohen IG, 2020, Geo L J, 108, 1425
- Muller H, 2021, Computer (Long Beach Calif), 54, 119
- Gao S, 2020, J Med Internet Res, 22
- Ye T, 2019, J Med Internet Res, 21
- American Medical Association. Code of Medical Ethics: Informed Consent & Shared Decision Making, https://www.ama-assn.org/delivering-care/ethics/informed-consent (accessed 10 December 2023).
- Schiff D, 2019, AMA J Ethics, 21
- Bond RR, 2018, J Electrocardiol, 51
Cited by
Weiner, E. B., Dankwa-Mullan, I., Nelson, W. A., & Hassanpour, S. (2025) · PLOS Digital Health · doi:10.1371/journal.pdig.0000810 · Cited by 218
Hoelscher, S. H., & Pugh, A. N. (2025) · Nursing Outlook · doi:10.1016/j.outlook.2025.102466 · Cited by 120
Patient Consent and The Right to Notice and Explanation of AI Systems Used in Health Care
Hurley, M. E., Lang, B. H., Kostick, K. M., Smith, J. N., & Blumenthal‐Barby, J. (2024) · The American Journal of Bioethics · doi:10.1080/15265161.2024.2399828 · Cited by 83
Olawade, D. B., Plabon, S. B., Ojo, A., Ogunbona, M. A., Makanjuola, B. D., & Olasilola, R. (2026) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2026.106362 · Cited by 105
Informed Consent for Ambient Documentation Using Generative AI in Ambulatory Care
Lawrence, K. S., Kuram, V., Levine, D. L., Sharif, S., Polet, C., Malhotra, K., & Owens, K. (2025) · JAMA Network Open · doi:10.1001/jamanetworkopen.2025.22400 · Cited by 82
Metadata
- Authors
- Park, H. J.
- Affiliations
- Hanyang University (KR)
- Corresponding author
- Park, Hai Jin
- ORCID
- Park, Hai Jin
- Venue
- DIGITAL HEALTH
- Publisher
- SAGE Publications
- Published
- 2024-01
- Volume
- 10
- Pages
- 20552076241247938-20552076241247938
- ISSN
- 2055-2076 (ISSN-L)
- PMC
- PMC11064747
- OpenAlex
- W4396557787
- Cited by
- 95
- References
- 42
- Citations by year
- 2026: 49 · 2025: 39 · 2024: 7
- FWCI
- 8.90
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Digital Health (publishedVersion, cc-by-nc-nd)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Funders
- Hanyang University (HY- 202100000002989)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Ethics and bioethics in healthcare
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Ethics in Clinical Research
- Keywords
- informed consent, AI-assisted diagnosis, clinical decision support, patient perspectives, AI disclosure, patient preferences, gender differences, age differences, income gap, AI ethics guidelines, radiology
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-04, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-04 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Patient perspectives on informed consent for medical AI: A web-based experiment. https://app.cochat.ai/papers/10.1177/20552076241247938
- The paper
- Park, H. J. (2024). Patient perspectives on informed consent for medical AI: A web-based experiment. DIGITAL HEALTH. https://doi.org/10.1177/20552076241247938
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.