‘What they’re not telling you about ChatGPT’: exploring the discourse of AI in UK news media headlines
Open accessAbstract
This study analyses the discursive representation of Artificial Intelligence (AI) and ChatGPT in UK news media headlines from January to May 2023. A total of 671 headlines were collected and analysed using inductive thematic analysis, theoretically informed by Agenda-Setting theory and Framing theory. The results offer an initial picture of how recent technological advances in the fields of AI have been communicated to the public. The results show that there is a complex and at times paradoxical portrayal of AI in general and ChatGPT as well as other Large Language Models (LLMs), oscillating between promising potential for solving societal challenges while simultaneously warning of imminent and systemic dangers. Further to this, the analysis provides evidence for the claim that media representations of AI are often sensationalised and tend to focus more on warnings and caution to readers, as only a minority of headlines were related to helpful, useful, or otherwise positive applications of AI, ChatGPT, and other Large Language Models (LLMs). These findings underscore the pivotal role of media discourse in shaping public perceptions of AI. The study prompts reflections on news media practices in the United Kingdom and encourages future research to further examine the influence of social, cultural, and political contexts on AI representation during a period of technological change. This research provides relevant insights for policymakers, AI developers, and educators to support public engagement with AI technologies.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
In a study of media headlines related to GenAI tools, the vast majority of headlines (including those on education) were concerned with the negative societal impacts of GenAI.
“In a study of media headlines related to GenAI tools, Roe and Perkins (2023) found that the vast majority of headlines (including those on education) were concerned with the negative societal impacts of GenAI”
From The Artificial Intelligence Assessment Scale (AIAS): A Framework for Ethical Integration of Generative AI in Educational Assessment (2024) · cites this paper as Roe and Perkins (2023)
The Results report that ‘Impending Danger’ was the largest theme and concerned potential societal damage, but it accounted for 248 of 671 headlines (37%), not a vast majority. Another 26% were ‘Explanation/Informative’, and the paper does not report a separate finding for education-related headlines. The study supports a tendency toward warnings, not the claimed proportion or education-specific conclusion. (LLM, conf=0.88)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 27 · Cited by 87
Research works printed but not linked: 20
- Aslam F, Awan TM, Syed JH et al. (2020) Sentiments and emotions evoked by news headlines of coronavirus disease (COVID-19) outbreak. Humanit Soc Sci Commun. 7:1-9. https://doi.org/10.1057/s41599-020-0523-3 · DOI
- Beelen K, Lawrence J, Wilson DCS, Beavan D (2023) Bias and representativeness in digitized newspaper collections: Introducing the environmental scan. Digit Scholarsh Humanit. 38:1-22. https://doi.org/10.1093/llc/fqac037 · DOI
- Blom JN, Hansen KR (2015) Click bait: Forward-reference as lure in online news headlines. J Pragmat. 76:87-100. https://doi.org/10.1016/j.pragma.2014.11.010 · DOI · Paper page
- Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, Neelakantan A, Shyam P, Sastry G, Askell A, Agarwal S, Herbert-Voss A, Krueger G, Henighan T, Child R, Ramesh A, Ziegler D, Wu J, Winter C, … Amodei D (2020) Language Models are Few-Shot Learners. Adv. Neural Inf Process Syst. 33:1877-1901. https://proceedings.neurips.cc/paper/2020/hash/ 1457c0d6bfcb4967418bfb8ac142f64a-Abstract.html
- Ghasiya P, Okamura K (2021) Understanding the Middle East through the eyes of Japan's Newspapers: a topic modelling and sentiment analysis approach. Digit Scholarsh Humanit. 36:871-885. https://doi.org/10.1093/llc/fqab019 · DOI
- Haider AS, Hussein RF (2020) Analysing headlines as a way of downsizing news corpora: Evidence from an Arabic and English comparable corpus of news- paper articles. Digit Scholarsh Humanit. 35:826-44. https://doi.org/10.1093/ llc/fqz074 · DOI
- Hassan A, Albayari A (2022) The usage of artificial intelligence in journalism. In: Hamdan A, Harraf A, Arora P, et al., (eds) Future of organizations and work after the 4th industrial revolution: the role of artificial intelligence, big data, automation, and robotics. Springer International Publishing, Cham, p 175-197
- Ifantidou E (2009) Newspaper headlines and relevance: ad hoc concepts in ad hoc contexts. J Pragmat. 41:699-720. https://doi.org/10.1016/j.pragma.2008.10.016 · DOI
- Jaillant L, Rees A (2023) Applying AI to digital archives: trust, collaboration and shared professional ethics. Digit Scholarsh Humanit. 38:571-585. https://doi. org/10.1093/llc/fqac073 · DOI
- Kuiken J, Schuth A, Spitters M, Marx M (2017) Effective headlines of newspaper articles in a digital environment. Digit J. 5:1300-1314. https://doi.org/10. 1080/21670811.2017.1279978 · DOI · Paper page
- Lecheler S, de Vreese CH (2018) News framing effects. Taylor & Francis Group, Milton MacRitchie V, Seedat M (2008) Headlines and discourses in newspaper reports on traffic accidents. South Afr J Psychol. 38:337-354. https://doi.org/10.1177/ 008124630803800206 · DOI
- Neuman R, Guggenheim L, Mo Jang S, Bae SY (2014) The dynamics of public attention: agenda-setting theory meets big data. J Commun. 64:193-214. https://doi.org/10.1111/jcom.12088 · DOI
- Perkins M, Roe J (2023) Decoding academic integrity policies: a corpus linguistics investigation of AI and other technological threats. High Educ Policy. https:// doi.org/10.1057/s41307-023-00323-2 · DOI
- Perkins M, Roe J, Postma D, et al (2023) Game of tones: faculty detection of GPT-4 generated content in university assessments. https://arxiv.org/abs/2305.18081 · arXiv
- Pichai, S (2023) Google AI updates: Bard and new AI features in Search. https:// blog.google/technology/ai/bard-google-ai-search-updates/. Accessed 23 May 2023a Roe J (2023) Discursive construction of contract cheating and degradation of higher education: comments on The Daily Mail online. Forum Qual Sozial- forschung/Forum: Qual Soc Res. 24. https://doi.org/10.17169/fqs-24.2.4000 · DOI
- Ross RM, Rand DG, Pennycook G (2021) Beyond "fake news": analytic thinking and the detection of false and hyperpartisan news headlines. Judgm Decis Mak. 16:484-504. https://doi.org/10.1017/S1930297500008640 · DOI
- Smelter TJ, Calvillo DP (2020) Pictures and repeated exposure increase perceived accuracy of news headlines. Appl Cogn Psychol. 34:1061-1071. https://doi. org/10.1002/acp.3684 · DOI · Paper page
- Veglis A, Maniou TA (2019) Chatbots on the rise: a new narrative in journalism. Stud Media Commun. 7:1-6. https://doi.org/10.11114/smc.v7i1.3986 · DOI
- Weber-Wulff D, Anohina-Naumeca A, Bjelobaba S, et al. (2023) Testing of detection tools for AI-generated text https://arxiv.org/abs/2306.15666 · arXiv
- Yoon L, Hernández D (2021) Energy, energy, read all about it: a thematic analysis of energy insecurity in the U.S. mainstream media from 1980 to 2019. Energy Res Soc Sc. 74:101972. https://doi.org/10.1016/j.erss.2021.101972 · DOI
- Authors
- Roe, J., & Perkins, M.
- Affiliations
- James Cook University Singapore (SG)
- British University Vietnam (VN)
- Corresponding author
- Roe, Jasper
- Venue
- Humanities and Social Sciences Communications
- Published
- 2023-10-30
- Volume
- 10
- Issue
- 1
- ISSN
- 2055-1045 (ISSN-L)
- 2662-9992
- OpenAlex
- W4388020713
- Cited by
- 87
- References
- 27
- Citations by year
- 2026: 28 · 2025: 31 · 2024: 26 · 2023: 2
- FWCI
- 41.27
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Humanities and Social Sciences Communications (publishedVersion, cc-by)
- ResearchOnline - JCU (James Cook University) (submittedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Durham Research Online (Durham University) (submittedVersion)
- Topics
- Social Sciences › Communication › Media Studies and Communication
- Social Sciences › Sociology and Political Science › Media Discourse and Social Analysis
- Social Sciences › Communication › Communication and COVID-19 Impact
- Keywords
- ChatGPT, artificial intelligence, news headlines, media representation of AI, large language models, agenda setting, framing theory, sensationalism, public perception of AI, British media
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.
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
