Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse
Open accessThe DOI matches its Crossref record; 89 of the 91 references it cites were found in a registry or an index; 4 of the 15 sources its claims cite were checked for support.
Abstract
Digital media and social networks offer adolescents a space for various social and sexual interactions. While the internet fosters healthy sexual exploration, it also facilitates online sexual abuse, such as non-consensual intimate image creation or dissemination, sextortion, sexual harassment, grooming, and gender-based hate speech. This abuse often goes underreported, especially when parents and teachers minimize or blame victims, or fail to recognize the abuse. This study examined how Israeli teachers and artificial intelligence (AI) identify and attribute blame in cases of sexual abuse. Six vignettes, depicting various sexual scenarios, were presented to 153 teachers and analyzed by ChatGPT-3.5 and ChatGPT-4. Findings revealed that teachers struggled to identify online sexual abuse and often blamed victims. ChatGPT-3.5 also tended to blame victims, but ChatGPT-4 accurately identified the abuse without victim-blaming. The study suggests GPT-4 can be an effective tool for identifying and naming sexual abuse when teachers lack proper training.
Overview
CoChat Summary
TL;DR
This study compared Israeli teachers with ChatGPT-3.5 and ChatGPT-4 in identifying, naming, and assigning blame in six vignettes about adolescent sexual scenarios. Teachers struggled to identify online sexual abuse and often blamed victims; GPT-3.5 showed similar victim-blaming, while GPT-4 accurately identified the abuse without blaming victims.
Key findings
- Teachers correctly named physical rape in 98% of cases but performed poorly on online abuse: sextortion 33.3%, non-consensual image sharing 60.1%, revenge-based image sharing 59.5%, online grooming 8.5%, and consensual sexting 23.5%. GPT-4 correctly named all vignettes at 100%; GPT-3.5 was mixed (e.g., sextortion 100%, revenge-based image sharing 100%, but sexting 20% and grooming 40%).
- Teachers had significant difficulty identifying online grooming as sexual abuse: only 16.3% said yes, 63.4% said no, and 20.3% were unsure. GPT-3.5 and GPT-4 both identified grooming as sexual abuse in 100% of responses. For consensual sexting, teachers said it was sexual abuse in 21.6% of cases, GPT-3.5 in 50%, and GPT-4 in 0%.
- Teachers assigned some responsibility to the victim in all online abuse cases: sextortion 78.4%, non-consensual image sharing 79.1%, online grooming 93.5%, and revenge-based image sharing 71.2%. GPT-4 assigned the victim responsibility in 0% of cases across these vignettes; GPT-3.5 blamed victims in 50% of sextortion, 90% of grooming, and 40% of revenge-based image-sharing responses.
- For physical rape, teachers assigned the victim some responsibility in 28.8% of cases, GPT-3.5 in 10%, and GPT-4 in 0%.
Methods
- Between December 2023 and February 2024, 153 teachers from across Israel were recruited via social media and WhatsApp groups and completed a questionnaire based on six vignettes: physical rape, consensual sexting, sextortion, non-consensual image distribution, revenge-based image distribution, and online grooming. All teacher participants were women; the sample was 69.3% Jewish and 30.7% Arab.
- For each vignette, participants answered three questions: labeling the event, judging whether it was sexual abuse, and assigning responsibility to victim and abuser on a 5-point Likert scale later dichotomized.
- In February 2024, GPT-3.5 and GPT-4 each received the same vignettes and questions 10 times in new chat windows. Chi-square tests compared teachers with the AI models using SPSS v27; tables report N = 173, reflecting 153 teachers plus 20 AI evaluations.
Limitations
- The paper states the sample was relatively small, recruited by convenience sampling, and all participants were women; future research should include larger diverse samples, male and non-binary educators, and analysis by religion or other identity factors.
- The paper notes all vignettes had boys as abusers and girls as victims, so future research should examine abuse against boys and non-binary youth by offenders of different genders. It also notes possible self-selection bias, since respondents were teachers willing to engage with sexual content, and that more conservative populations might show lower naming ability and higher victim-blaming.
- Evident from the text: AI performance was based on only 10 measurements per vignette per model, and the study used vignettes developed specifically for this research rather than a broader validated stimulus set.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures






Verification
Claims the paper makes
4 Valid · 0 Not valid · 6 Not checked · 48 Own results and interpretations
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
“Studies have shown that educators often find it difficult to report their suspicions regarding sexual abuse because of a lack of knowledge about the abuse or the required procedure for handling it (Lindenbach et al., 2021;Mathews, 2011;Rochford et al., 2023;Sigad et al., 2023;Zinn et al., 2024).”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 1.2. Teachers' responses to online sexual abuse: barriers and biases
What the sources say
- Not checked — no full text of the sourcecites Lindenbach et al., 2021 source
- Supportedcites Mathews, 2011 source
The “Existing Research into Teacher Training” section reports that teachers without recent training had significant gaps in knowledge of their reporting duty and that a national study found many lacked sufficient knowledge to detect and report suspected cases. It also says training and confidence in recognising abuse influence reporting practice, supporting the claimed link between knowledge gaps and difficulty reporting suspicions. (LLM, conf=0.89)
- Not checked — no full text of the sourcecites Rochford et al., 2023 source
boostedstates a cause- 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.
Every source the paper cites for this claim was checked and supports it.
In the paper
“Teachers are often unaware or ill-equipped to address the issue, especially when it involves harmful sexual behaviors perpetrated by children and adolescents (Kor et al., 2023).”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 1.2. Teachers' responses to online sexual abuse: barriers and biases
What the sources say
- Supportedcites Kor et al., 2023 source
The Introduction (‘Harmful Sexual Behavior in Schools’; page not provided) reports that teachers were unaware of the scale of students’ harmful sexual behavior or ill-prepared to respond to it. The ‘Minimization of HSB’ section (page not provided) adds that online-safety training was lacking and that frontline teachers felt unprepared to identify harmful sexual behavior and support affected students. (LLM, conf=0.88)
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
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
“Findings revealed that teachers struggled to identify online sexual abuse and often blamed victims.”
- Also stated in: § 4. Discussion · boosted
“The study found overall that teachers had great difficulty naming online sexual abuse.”
- Also stated in: § 5. Conclusion · plain
“The findings underscore the challenges faced by teachers in recognizing and accurately naming various forms of sexual abuse, especially in the online context.”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > Abstract
boosted- 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: § 4. Discussion · boosted
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-3.5 also tended to blame victims, but ChatGPT-4 accurately identified the abuse without victim-blaming.”
- Also stated in: § 4. Discussion · hedged
“GPT-3.5 adopted the same position and appeared to share the same biases in blaming the victims.”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > Abstract
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.
- Also stated in: § 4. Discussion · hedged
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
“The findings revealed significant differences between the three groups of participants in identifying sexual abuse in cases of sexting and online sexual grooming.”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 3.2. Identification of the incident as sexual abuse
plainstates an association- 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.
30 more claims
Create a free account to see all 58 claims and to verify the 3 references still open.
- Probably not a substantial claim. 13
- It may state two separate claims as one. 16
- The quote only partly states this claim. 2
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 major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 4. Discussion
“The present study found a significant difference between GPT-4 and teachers in the measure of the victim's responsibility for the rape.”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 3.3. Victim blaming
“The study found significant differences between the three groups in identifying responsibility in the case of online sexual abuse, but no differences in the vignettes of rape in the physical space.”
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
“| Secular | 75 (43.4%) |”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 6. Limitations and future directions
Reported 43.4 · recomputed 49.019608
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-3.5 | 6 (60%) | 4 (4%) | (.61) |”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 6. Limitations and future directions
Reported 4 · recomputed 40
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
“| Religious | 21 (12.1%) |”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 6. Limitations and future directions
Reported 12.1 · recomputed 13.72549
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
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 2.2.2. Sexual violence
“After reading each vignette, participants were asked to answer Yes or No to the question: 'Is the described event a case of sexual abuse?'”
§ Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse > 6. Limitations and future directions
“| Sexting | Teachers | 77 (50.3%) | 33 (21.6%) | 43 (28/1%) | 15.78** |”
1 more finding
Create a free account to see every finding of this check.
Citations
Cites
Social Change and Crime Rate Trends: A Routine Activity Approach
Cohen, L. E., & Felson, M. (1979) · American Sociological Review · doi:10.2307/2094589 · Cited by 8,856
Just world research and the attribution process: Looking back and ahead.
Lerner, M. J., & Miller, D. T. (1978) · Psychological Bulletin · doi:10.1037/0033-2909.85.5.1030 · Cited by 1,687 · cited 1×
“The just-world hypothesis (Lerner & Miller, 1978) argues that individuals seek to find logical and just explanations for events, leading them to blame victims as a way of maintaining the belief that bad things happen to those who engage in behavior that invites harm.”
Technology-facilitated sexual violence: A literature review of empirical research
Henry, N., & Powell, A. (2018) · Trauma, Violence, & Abuse · doi:10.1177/1524838016650189 · Cited by 675 · cited 1×
“…(Schoeps et al., 2020), or where the perpetrator threatens to share intimate images of the victim to blackmail them into sending more images, paying money, performing a sexual act, or preventing them from breaking up the relationship (‘sextortion’) (Henry & Powell, 2018; Henry & Umbach, 2024).”
Self-citationTeen girls, sexual double standards and ‘sexting’: Gendered value in digital image exchange
Ringrose, J., Harvey, L., Gill, R., & Livingstone, S. (2013) · Feminist Theory · doi:10.1177/1464700113499853 · Cited by 618 · cited 1×
“…a double sexual standard, whereby abused girls are accused of losing their dignity, whereas the boys who distribute the images are often recognized and valued for their masculinity and sexual prowess (Dolev-Cohen, 2024; Ricciardelli & Adorjan, 2019; Ringrose et al., 2013; Ring-rose et al., 2021).”
ChatGPT makes medicine easy to swallow: an exploratory case study on simplified radiology reports
Jeblick, K., Schachtner, B., Dexl, J., Mittermeier, A., Stüber, A. T., Topalis, J., Weber, T., Wesp, P., Sabel, B. O., Ricke, J., et al. (2023) · European Radiology · doi:10.1007/s00330-023-10213-1 · Cited by 549 · cited 1×
“Both versions are popular and remarkably successful so that ChatGPT has great potential to become a daily assistant for purposes of general or professional consulting (Jeblick et al., 2023).”
Prevalence of Multiple Forms of Sexting Behavior Among Youth: A Systematic Review and Meta-analysis
Madigan, S., Ly, A., Rash, C. L., Van Ouytsel, J., & Temple, J. R. (2018) · JAMA Pediatrics · doi:10.1001/jamapediatrics.2017.5314 · Cited by 519 · cited 2×
“A meta-analysis conducted by Madigan et al. (2018) shows that 8.4% of adolescents reported that sexual images of themselves were distributed without consent, and another study found that 63.9% indicated that subjects had passed on sexting messages without the original sender’s consent…”
InfluentialHow Close is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detection
Guo, B., Zhang, X., Wang, Z., Jiang, M., Nie, J., Ding, Y., Yue, J., & Wu, Y. (2023) · arXiv · doi:10.48550/arxiv.2301.07597 · Cited by 318 · cited 1×
“It understands and interprets simple and complex user queries and responds in natural language (Alberts et al., 2023; Guo et al., 2023).”
Kahn, A. S., Jackson, J., Kully, C., Badger, K., & Halvorsen, J. (2003) · Psychology of Women Quarterly · doi:10.1111/1471-6402.00103 · Cited by 278 · cited 1×
“The vignette described stereotypical rape characteristics (i.e., the perpetrator was a stranger and used force, and the victim resisted), which is easier to identify than other rape scenarios (Hammond & Calhoun, 2007; Kahn et al., 1994; Kahn et al., 2003).”
Sextortion of Minors: Characteristics and Dynamics
Wolak, J., Finkelhor, D., Walsh, W., & Treitman, L. (2018) · Journal of Adolescent Health · doi:10.1016/j.jadohealth.2017.08.014 · Cited by 249 · cited 2×
“…several possible explanations for the underreporting of online sexual abuse, including the adolescents’ age, feelings of shame and guilt about the event, attitudes toward receiving help, fear of further harm, and the belief that they cannot be helped (Tomczyk & Kopecký, 2016; Wolak et al., 2018).”
Let’s talk about sexting, baby: Computer-mediated sexual behaviors among young adults
Drouin, M., Vogel, K. N., Surbey, A., & Stills, J. R. (2013) · Computers in Human Behavior · doi:10.1016/j.chb.2012.12.030 · Cited by 241 · cited 1×
“However, Sexting is not necessarily a risky behavior when it is part of a reciprocal and consensual process of flirting and creating intimacy (Bianchi et al., 2021; Dir et al., 2013; Dolev-Cohen, 2023; Drouin et al., 2013; Walker et al., 2013).”
Research works printed but not linked: 1
- Ezioni, L. (2020). The crime of grooming. Child and Family Law Journal, 8(1), 1–18. https://lawpublications.barry.edu/cgi/viewcontent.cgi?article=1041&context=cflj
Cited by
Dolev-Cohen, M., & Shaul, H. (2025) · Computers in Human Behavior · doi:10.1016/j.chb.2025.108777 · Cited by 14
Self-citationEfrati, Y. (2025) · Children and Youth Services Review · doi:10.1016/j.childyouth.2025.108477 · Cited by 1
Perceptions of Adolescent Sexting: A Cross-Cultural Study of Israeli and Italian Parents
Dolev-Cohen, M., Ricon, T., Colombo, C. C., & Cuccì, G. (2026) · Sexuality & Culture · doi:10.1007/s12119-026-10569-8 · Cited by 1
Self-citationMarcantonio, T. L., Vowels, L. M., LeRoux, S., Bravo, M., Thrash, A., Mueller, A., Niolet, A., Hall, M., Döring, N. M., & Nadarzynski, T. (2026) · The Journal of Sex Research · doi:10.1080/00224499.2026.2686406 · Cited by 1
Dolev-Cohen, M., & Shaul, H. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1790264 · Cited by 1
Self-citation
Metadata
- Authors
- Dolev-Cohen, M., & Henry, N.
- Affiliations
- Oranim Academic College of Education (IL)
- RMIT University (AU)
- Corresponding author
- Dolev-Cohen, Michal
- Venue
- Information, Communication & Society
- Publisher
- Informa UK Limited
- Published
- 2025-04-17
- Volume
- 28
- Issue
- 16
- Pages
- 2976-2994
- ISSN
- 1369-118X (ISSN-L)
- 1468-4462
- OpenAlex
- W4409531005
- Semantic Scholar
- 93025f746a08e23329d6cc3e6a8bd2e1977ba534
- Cited by
- 8
- References
- 91
- Citations by year
- 2026: 6 · 2025: 2
- FWCI
- 14.58
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Information Communication & Society (publishedVersion, cc-by-nc-nd)
- Topics
- Social Sciences › Gender Studies › Sexual Assault and Victimization Studies
- Social Sciences › Education › Artificial Intelligence in Education
- Social Sciences › Psychology › Clinical Psychology › Child Abuse and Trauma
- Keywords
- online sexual abuse, teachers, ChatGPT, blame attribution, victim blaming, sextortion, sexual harassment, online grooming, gender-based hate speech, teacher professional development, adolescents, Israeli teachers
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 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (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: Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse. https://app.cochat.ai/papers/10.1080/1369118x.2025.2492585
- The paper
- Dolev-Cohen, M., & Henry, N. (2025). Teachers vs. Chat GPT: identifying and naming acts of online sexual abuse. Information, Communication & Society. https://doi.org/10.1080/1369118x.2025.2492585
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.