Pith. sign in

REVIEW 4 major objections 4 minor 67 references

Large Language Models as 'Hidden Persuaders': Fake Product Reviews are Indistinguishable to Humans and Machines

T0 review · 4 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read People and frontier AI models both fail to distinguish real from machine-written product reviews, three studies report.

desk verdict A careful but narrow benchmark shows humans and LLMs both miss well-crafted fake reviews; the abstract's broad claim outruns the evidence, but the data and analysis deserve peer review. read the letter →

arxiv 2506.13313 v1 pith:RTWU7KYR submitted 2025-06-16 cs.CL cs.AIecon.GNq-fin.EC

classification cs.CLcs.AIecon.GNq-fin.EC
keywords fakeproductreviewsLLM-generatedtextAIcontentdetectionconsumertrustonlinereviewauthenticityhuman-AIcomparisonscepticismbiasveracity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Three studies report that people and large language models can no longer reliably tell real product reviews from machine-written ones. In a task with 25 real and 25 ChatGPT-o1-generated reviews, 288 human judges averaged 50.8% accuracy, essentially chance, despite moderate confidence; seven frontier LLMs scored 35.6–50.0%, all at or below the human rate and most detecting only a small share of fakes. The paper explains the failures through two opposing heuristics: humans display a 'scepticism bias' that treats polished, highly positive reviews as fake, while LLMs display a 'veracity bias' that defaults to believing reviews are real and relies on shallow cues such as length. If this is right, text-based authenticity checks by readers or general-purpose AI are no longer a dependable safeguard, and online review systems need verification of purchase or authorship rather than detection.

What carries the argument

The mechanism that carries the argument is the construction and comparison of the 50-review stimulus set. The authors first profile a thousand real online marketplace reviews for length, punctuation, pronoun use, past-tense verbs, idioms, mistakes, and sentiment; they convert those patterns into a prompt for ChatGPT-o1 so that the fake reviews reproduce human imperfections at a targeted rate (five mistakes per ten reviews, concentrated in mid- and low-star reviews). They then run the same 50 reviews through 288 human judges and seven LLMs under matched instructions, and use class-wise precision, recall, F1, and a cosine similarity between human and model judgment vectors to distinguish shared cues from divergent ones. Named explanatory constructs—humans' 'scepticism bias' toward too-good-to-be-true reviews and LLMs' 'veracity bias' toward accepting text as real—do the interpretive work in Study 3.

What would settle it

Generate fake reviews with several different LLMs and with no injected typos, then rerun the same 50-review classification task; if human accuracy rises clearly above chance (for example, above 60%) or any LLM detector reaches high accuracy, the paper's indistinguishability claim is specific to its generation recipe rather than to LLM-generated reviews generally.

Watch

Extended reading notes

Core claim

At the paper's core is the finding that LLM-generated fake product reviews are indistinguishable from authentic human reviews to both people and machines. The evidence is a 50-review benchmark: 25 real reviews sampled from a public marketplace corpus and 25 fake reviews written by ChatGPT-o1 under prompts built from the corpus's observed style, including deliberate misspellings and grammar slips (about five errors per ten reviews). In the human study, accuracy was 50.82%; participants recognized 65.8% of real reviews but only 35.8% of fakes. In the LLM study, the best model matched human accuracy at 50.0% and the worst scored 35.6%, with LLMs as a group identifying only about 9% of fake reviews. The third study attributes the near-chance performance to systematic, opposed biases: human judges are skeptical of overly positive and polished reviews, while LLM judges are biased toward believing reviews are real and lean on surface richness such as length.

Load-bearing premise

The result hinges on the premise that the 25 ChatGPT-o1 reviews written from these specific prompts—which deliberately add about five errors per ten reviews—stand in for the general class of LLM-generated fake reviews in the wild; other models, prompt styles, or error rates could change human and machine detection rates substantially.

Editorial extensions

If this is right

  • Review platforms that rely on human flagging or general LLM screening cannot expect to catch machine-written fakes; verified-purchase and provenance signals become the only dependable defense.
  • Consumers' scepticism bias means polished, highly positive genuine reviews will keep getting dismissed as fake, while fake negative reviews—a cheap way to damage competitors—are especially likely to be believed.
  • General-purpose LLMs are not a valid detector baseline for this task; any moderation pipeline using them should assume near-zero recall for fake reviews unless reweighted or fine-tuned.
  • In unbalanced real-world settings, where most reviews are genuine, the LLM default of 'real' will appear accurate overall while letting essentially all fakes through, making prevalence-based metrics dangerously misleading.
  • If indistinguishability holds across product categories, consumer research that treats online review text as ground-truth human opinion must contend with AI pollution of its data.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: A direct test of the generation recipe—producing fake reviews with zero injected mistakes or with several other LLMs—would show whether the near-chance result is a property of LLM text itself or partly an artifact of the authors' deliberate imperfection schedule.
  • Editorial inference: The paper's asymmetry (humans miss fakes, LLMs call everything real) suggests a combined human-machine pipeline would still have low recall for fakes; what is needed is calibration against known base rates rather than another detector.
  • Editorial inference: If the veracity bias reflects training-data priors, then detection prompts that force 'fake unless proven otherwise' or give class-balanced instructions could shift LLM behavior; the paper used a single standard prompt, so prompt sensitivity remains open.
  • Editorial inference: The finding implies watermarking or metadata disclosure at generation time is the more scalable intervention, because the paper's own data confirm that post-hoc text inspection by either reader or machine is at chance.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper investigates whether humans and large language models can distinguish real Amazon product reviews from fake reviews generated by ChatGPT-o1 under a heuristic-guided prompting protocol. Study 1 reports that 288 human participants judged 25 real and 25 fake reviews with an overall accuracy of 50.82%, which the authors describe as chance-level. Study 2 benchmarks seven LLMs on the same 50 reviews, reporting accuracies between 35.6% and 50.0%, and claims that humans slightly outperform all tested LLMs. Study 3 correlates review features with classification difficulty and introduces the constructs of human 'scepticism bias' and LLM 'veracity bias'. The abstract concludes that humans and machines can no longer distinguish fake from real product reviews and that review systems everywhere are now susceptible to mechanised fraud.

Significance. If the results hold, this would be a valuable benchmark for human and LLM detection of adversarially matched product reviews, and the class-specific confusion patterns are interesting in their own right. The paper has real strengths: fake reviews were generated from explicit heuristics grounded in real-review statistics; the real and fake sets were compared on multiple distributional features; LLM judgments were repeated three times; and the data are openly available. The main contributions, however, depend on two things the current manuscript does not supply: proper inferential statistics for the chance-level and human-vs-LLM claims, and a scoping of the conclusions to the single-generator, single-prompt condition actually studied.

major comments (4)
  1. [§3.4.1 / Abstract] The headline claim that humans perform 'essentially the same as chance' at 50.82% is reported without any inferential statistic. Because accuracy is averaged over 288 participants, a one-sample test or 95% confidence interval is needed; with the reported per-participant SD of 0.08, the interval will be informative, and 'compatible with chance' is not the same as 'no better than chance' unless a formal test is reported. Please add a test that accounts for participant and review clustering (e.g., a mixed-effects logistic regression) and condition the abstract on the result.
  2. [§4.3, Table 5] The comparison between human accuracy (50.82%) and LLM accuracy (35.6%-50.0%) is purely descriptive; no significance test or confidence interval is given, and the LLM estimates rest on only 50 reviews with three repetitions per model. The claim that humans 'slightly outperformed all tested LLMs' is therefore not established. Report a test that accounts for repeated LLM trials and participant clustering, or explicitly label the cross-system ranking as exploratory.
  3. [§3.2, §7.2] The abstract and title generalize beyond the evidence: the entire stimulus set uses 25 fake reviews from a single LLM (ChatGPT-o1) under one hand-crafted prompt with explicit error injection and distribution matching. Section 7.2 acknowledges the single-LLM limitation, but the manuscript still asserts as a general fact that fake product reviews are indistinguishable to humans and machines. This single-point result cannot support 'review systems everywhere are now susceptible.' The claims should either be narrowed to the specific generation pipeline or the study should be extended to multiple generators, prompts, and product categories; note also that ChatGPT-o1 itself is included among the detectors, so the measured near-chance performance may partly reflect a specific generator-detector pair.
  4. [§5.2 / Table 3] Study 3 reports many Spearman correlations on n=50 reviews without multiple-comparison correction, so the specific heuristic conclusions (e.g., LLMs rely on word count, humans show a scepticism bias) are at risk of false positives. In addition, the KS tests in Table 3 with n=25 per group cannot 'confirm' the absence of distributional differences; the descriptive means differ substantially on word count (35.92 vs 66.44) and helpful votes (0.52 vs 1.92). Report equivalence tests or effect sizes with confidence intervals, and apply a multiple-comparison correction in Study 3.
minor comments (4)
  1. [Table 5] The precision, recall, and F1 columns appear to be macro-averaged across the real and fake classes, but this is not stated; under the standard non-macro definition the values are inconsistent, e.g., ChatGPT-o1 with precision 0.752 and recall 0.507 would give F1 about 0.605, not 0.348. Please define the averaging scheme in the table note.
  2. [§3.3.3] The confidence scale is not defined in the measures subsection; please state the response scale (e.g., 0-100) and the exact wording used for the confidence question.
  3. [Throughout] There are typographical errors that should be corrected in revision, including 'utilsed' in Section 3, 'heuristicsh' in the Table 2 header, and 'V ADER' in Table 1.
  4. [§5.1] Grok-3's exclusion is justified by its constant 'real' response, but this constant-response behavior is itself a substantive finding under the paper's 'veracity bias' interpretation; consider analyzing it explicitly as a degenerate case rather than simply dropping it from Study 3.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: accuracy results are measured, not fitted; minor self-citation and prompt-design caveats do not make the findings tautological.

full rationale

The paper's central results are experimental measurements: 288 human participants classified 50 reviews at 50.82% accuracy, and seven LLMs scored at or below chance (Tables 5 and 6). These numbers are not derived from the generation parameters; no parameter is fitted to the human or LLM accuracy outcomes. The fake reviews were produced by a single LLM (ChatGPT-o1) under prompt heuristics (Table 2) derived from 1,000 real Home_and_Kitchen reviews (Section 3.1), and Table 3 shows the resulting fakes match real reviews on the measured features. That is a calibration check, not a circular reduction: the classification task uses the same 50 stimuli but independent judges, and the chance-level outcomes are contingent empirical results. The self-referential elements (the same model family generated and judged; the prompt includes 'The aim should be that a reader cannot distinguish the reviews from standard writing') are generalization and construct-validity concerns, acknowledged in Section 7.2, rather than reductions of the measured accuracies to the generation inputs. The only self-citation (Meng, 2025, for the 50-review dataset) is not load-bearing because the dataset is fully described in Sections 3.2 and 3.3.1. No specific circular step can be exhibited, so no circularity is flagged.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claim depends less on fitted parameters than on design choices. The only hand-set numeric value is the error rate injected into fake reviews. The axioms concern the representativeness of the Amazon sample, the representativeness of ChatGPT-o1 as a stand-in for LLM review fraud, the use of default LLM settings, and the sufficiency of a 50-review stimulus set. No new physical or mathematical entities are introduced.

free parameters (1)
  • Error injection rate in fake review prompts = 5 mistakes per 10 reviews
    Chosen by hand in the generation prompt (Table 2) to match the imperfection rate observed in real Home & Kitchen reviews. The indistinguishability result depends on such calibration; a different rate would likely change detection accuracy.
assumptions (4)
  • domain assumption Amazon Review 2023 Home & Kitchen reviews are representative of real human-written product reviews.
    The ground truth real class is taken from the dataset without manual verification that each of the 25 sampled reviews is genuine consumer text.
  • domain assumption ChatGPT-o1 with the specified prompts produces fake reviews representative of the broader class of LLM-generated review fraud.
    The paper uses a single generation model and a single prompt set; generalization to other LLMs and prompt strategies is assumed in the abstract's broad wording.
  • domain assumption LLM API responses under default settings reflect the models' typical detection ability.
    Each model was queried three times with default temperature; no attempt was made to optimize prompts for detection, and the paper acknowledges this in Section 7.2.
  • domain assumption 50 reviews (25 per class) are sufficient to estimate human and LLM accuracy stably.
    The small stimulus set constrains the correlation analyses in Study 3 and the precision of accuracy estimates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Large Language Models as 'Hidden Persuaders': Fake Product Reviews are Indistinguishable to Humans and Machines." pith.science (2026). https://pith.science/paper/RTWU7KYR

@misc{pith2026250613313,
  author       = {Pith},
  title        = {Pith review of: Large Language Models as 'Hidden Persuaders': Fake Product Reviews are Indistinguishable to Humans and Machines},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RTWU7KYR}},
  note         = {Machine review of arXiv:2506.13313}
}
read the original abstract

Reading and evaluating product reviews is central to how most people decide what to buy and consume online. However, the recent emergence of Large Language Models and Generative Artificial Intelligence now means writing fraudulent or fake reviews is potentially easier than ever. Through three studies we demonstrate that (1) humans are no longer able to distinguish between real and fake product reviews generated by machines, averaging only 50.8% accuracy overall - essentially the same that would be expected by chance alone; (2) that LLMs are likewise unable to distinguish between fake and real reviews and perform equivalently bad or even worse than humans; and (3) that humans and LLMs pursue different strategies for evaluating authenticity which lead to equivalently bad accuracy, but different precision, recall and F1 scores - indicating they perform worse at different aspects of judgment. The results reveal that review systems everywhere are now susceptible to mechanised fraud if they do not depend on trustworthy purchase verification to guarantee the authenticity of reviewers. Furthermore, the results provide insight into the consumer psychology of how humans judge authenticity, demonstrating there is an inherent 'scepticism bias' towards positive reviews and a special vulnerability to misjudge the authenticity of fake negative reviews. Additionally, results provide a first insight into the 'machine psychology' of judging fake reviews, revealing that the strategies LLMs take to evaluate authenticity radically differ from humans, in ways that are equally wrong in terms of accuracy, but different in their misjudgments.

Figures

Figures reproduced from arXiv: 2506.13313 by the authors.

Figure 1
Figure 1. Comparison of confusion matrices for human participants and LLMs. [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. Relationship between participant age and classification accuracy, with a fitted quadratic trendline to illustrate [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

67 extracted references · 64 canonical work pages

  1. [1]

    M., Del Barrio-García, S., Porcu, L., and Crespo-Almendros, E

    Alcántara-Pilar, J. M., Del Barrio-García, S., Porcu, L., and Crespo-Almendros, E. (2017). Language as a cultural vessel in online servicescapes: Its impact on consumers’ perceived risk, attitudes, and behavioural intentions.Journal of Consumer Behaviour, 16(6):e61–e75

  2. [2]

    and McFarlane, S

    Alkaissi, H. and McFarlane, S. I. (2023). Artificial hallucinations in chatgpt: implications in scientific writing.Cureus, 15(2)

  3. [3]

    Y ., and Kim, J.-J

    Banerjee, S., Chua, A. Y ., and Kim, J.-J. (2017). Don’t be deceived: Using linguistic analysis to learn how to discern online review authenticity.Journal of the Association for Information Science and Technology, 68(6):1525–1538

  4. [4]

    A., Jafari-Sadeghi, V ., and Sakka, G

    Beheshti, M., ZareRavasan, A., Mahdiraji, H. A., Jafari-Sadeghi, V ., and Sakka, G. (2024). An overview of the consumer-centric disruptive technology research: Insights from topic modelling and literature review.Journal of Consumer Behaviour, 23(2):372–388

  5. [5]

    Bluvstein, S., Zhao, X., Barasch, A., and Schroeder, J. (2024). Imperfectly human: The humanizing potential of (corrected) errors in text-based communication.Journal of the Association for Consumer Research, 9(3):332–343

  6. [6]

    (2017).Superintelligence

    Bostrom, N. (2017).Superintelligence. Dunod

  7. [7]

    M., Friestad, M., and Rose, G

    Boush, D. M., Friestad, M., and Rose, G. M. (1994). Adolescent skepticism toward tv advertising and knowledge of advertiser tactics.Journal of consumer research, 21(1):165–175

  8. [8]

    A., Olson, A., Morgan, D

    Freed, J. A., Olson, A., Morgan, D. J., Hom, J., Gallo, R., Horvitz, E., Chen, J., Manrai, A. K., and Rodman, A. (2024). Superhuman performance of a large language model on the reasoning tasks of a physician

Show all 67 references
  1. [9]

    T., and Baykal, N

    Canbek, G., Sagiroglu, S., Temizel, T. T., and Baykal, N. (2017). Binary classification performance measures/metrics: A comprehensive visualized roadmap to gain new insights. In2017 International Conference on Computer Science and Engineering (UBMK), pages 821–826. IEEE

  2. [10]

    Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion.Journal of personality and social psychology, 39(5):752

  3. [11]

    Chevalier, J. A. and Mayzlin, D. (2006). The effect of word of mouth on sales: Online book reviews.Journal of marketing research, 43(3):345–354

  4. [12]

    N., Li, T., Li, D., Zhu, B., Zhang, H., Jordan, M., Gonzalez, J

    Chiang, W.-L., Zheng, L., Sheng, Y ., Angelopoulos, A. N., Li, T., Li, D., Zhu, B., Zhang, H., Jordan, M., Gonzalez, J. E., et al. (2024). Chatbot arena: An open platform for evaluating llms by human preference. InForty-first International Conference on Machine Learning. 18

  5. [13]

    Clark, E., August, T., Serrano, S., Haduong, N., Gururangan, S., and Smith, N. A. (2021). All that’s’ human’is not gold: Evaluating human evaluation of generated text.arXiv preprint arXiv:2107.00061

  6. [14]

    N., Japkowicz, N., and Viktor, H

    Crothers, E. N., Japkowicz, N., and Viktor, H. L. (2023). Machine-generated text: A comprehensive survey of threat models and detection methods.IEEE Access, 11:70977–71002

  7. [15]

    G., van Esch, P., and Phelan, S

    Cui, Y . G., van Esch, P., and Phelan, S. (2024). How to build a competitive advantage for your brand using generative ai.Business Horizons

  8. [16]

    G., Broda, E., Lindgren, E., Strömbäck, J., Tsfati, Y ., and Vliegenthart, R

    Damstra, A., Boomgaarden, H. G., Broda, E., Lindgren, E., Strömbäck, J., Tsfati, Y ., and Vliegenthart, R. (2021). What does fake look like? a review of the literature on intentional deception in the news and on social media.Journalism Studies, 22(14):1947–1963

  9. [17]

    Matejovicova, T., et al. (2024). Scalable watermarking for identifying large language model outputs.Nature, 634(8035):818–823

  10. [18]

    and Geere BA, L

    Djafarova, E. and Geere BA, L. (2023). Exploring the perceived online review credibility and management response influence on purchase intention.Journal of Information Technology Management, 15(2):92–115

  11. [19]

    Doctorow, C. (2023). Tiktok’s enshittification.Pluralistic: Daily Links from Cory Doctorow

  12. [20]

    and Hwang, J.-S

    Doh, S.-J. and Hwang, J.-S. (2009). How consumers evaluate ewom (electronic word-of-mouth) messages.Cyberpsy- chology & behavior, 12(2):193–197

  13. [21]

    Double, K. S. and Birney, D. P. (2024). Confidence judgments interfere with perceptual decision making.Scientific Reports, 14(1):14133

  14. [22]

    Duan, W., Gu, B., and Whinston, A. B. (2008). Do online reviews matter?—an empirical investigation of panel data. Decision support systems, 45(4):1007–1016

  15. [23]

    Amazeen, M. A. (2022). The psychological drivers of misinformation belief and its resistance to correction.Nature Reviews Psychology, 1(1):13–29

  16. [24]

    Ferraro, C., Demsar, V ., Sands, S., Restrepo, M., and Campbell, C. (2024). The paradoxes of generative ai-enabled customer service: A guide for managers.Business Horizons

  17. [25]

    J., Soohoo, C., Danielson, D

    Fogg, B. J., Soohoo, C., Danielson, D. R., Marable, L., Stanford, J., and Tauber, E. R. (2003). How do users evaluate the credibility of web sites? a study with over 2,500 participants. InProceedings of the 2003 conference on Designing for user experiences, pages 1–15

  18. [26]

    Ford, J., Jain, V ., Wadhwani, K., and Gupta, D. G. (2023). Ai advertising: An overview and guidelines.Journal of Business Research, 166:114124

  19. [27]

    Forman, C., Ghose, A., and Wiesenfeld, B. (2008). Examining the relationship between reviews and sales: The role of reviewer identity disclosure in electronic markets.Information systems research, 19(3):291–313

  20. [28]

    R., McCarthy, I

    Hannigan, T. R., McCarthy, I. P., and Spicer, A. (2024). Beware of botshit: How to manage the epistemic risks of generative chatbots.Business Horizons, 67(5):471–486

  21. [29]

    R., et al

    Esplin, C. R., et al. (2025). When eliza meets therapists: A turing test for the heart and mind.PLOS Mental Health, 2(2):e0000145

  22. [30]

    Hou, Y ., Li, J., He, Z., Yan, A., Chen, X., and McAuley, J. (2024). Bridging language and items for retrieval and recommendation.arXiv preprint arXiv:2403.03952

  23. [31]

    Hu, N., Liu, L., and Zhang, J. J. (2008). Do online reviews affect product sales? the role of reviewer characteristics and temporal effects.Information Technology and management, 9:201–214

  24. [32]

    B., Ballouk, H., Arfi, W

    Jabeur, S. B., Ballouk, H., Arfi, W. B., and Sahut, J.-M. (2023). Artificial intelligence applications in fake review detection: Bibliometric analysis and future avenues for research.Journal of Business Research, 158:113631

  25. [33]

    T., and Naaman, M

    Jakesch, M., Hancock, J. T., and Naaman, M. (2023). Human heuristics for ai-generated language are flawed. Proceedings of the National Academy of Sciences, 120(11):e2208839120

  26. [34]

    J., Chi, O

    Jia, S. J., Chi, O. H., and Chi, C. G. (2025). Unpacking the impact of ai vs. human-generated review summary on hotel booking intentions.International Journal of Hospitality Management, 126:104030

  27. [35]

    John, D. R. (1999). Consumer socialization of children: A retrospective look at twenty-five years of research.Journal of consumer research, 26(3):183–213. 19 Kovács, B. (2024). The turing test of online reviews: Can we tell the difference between human-written and gpt-4-writte...

  28. [36]

    Kusumasondjaja, S., Shanka, T., and Marchegiani, C. (2012). Credibility of online reviews and initial trust: The roles of reviewer’s identity and review valence.Journal of Vacation Marketing, 18(3):185–195

  29. [37]

    Lang, A. (2000). The limited capacity model of mediated message processing.Journal of communication, 50(1):46–70

  30. [38]

    and Park, J.-S

    Lee, M. and Park, J.-S. (2022). Do parasocial relationships and the quality of communication with ai shopping chatbots determine middle-aged women consumers’ continuance usage intentions?Journal of Consumer Behaviour, 21(4):842–854

  31. [39]

    Levin, I. P. and Gaeth, G. J. (1988). How consumers are affected by the framing of attribute information before and after consuming the product.Journal of consumer research, 15(3):374–378

  32. [40]

    Li, B., Yao, R., and Nan, Y . (2023). How do friendship artificial intelligence chatbots (faic) benefit the continuance using intention and customer engagement?Journal of Consumer Behaviour, 22(6):1376–1398

  33. [41]

    and Fleming, P

    Lindebaum, D. and Fleming, P. (2024). Chatgpt undermines human reflexivity, scientific responsibility and responsible management research.British Journal of Management, 35(2):566–575

  34. [42]

    Litwin, P., Paulewicz, B., and Siedlecka, M. (2025). Reporting confidence decreases response and change-of-mind accuracy in a perceptual decision task.Journal of Experimental Psychology: Human Perception and Performance

  35. [43]

    Liu, Y ., Zhang, Z., Zhang, W., Yue, S., Zhao, X., Cheng, X., Zhang, Y ., and Hu, H. (2023). Argugpt: evaluating, understanding and identifying argumentative essays generated by gpt models.arXiv preprint arXiv:2304.07666

  36. [44]

    C., and Bernritter, S

    Maslowska, E., Malthouse, E. C., and Bernritter, S. F. (2017). Too good to be true: the role of online reviews’ features in probability to buy.International Journal of Advertising, 36(1):142–163

  37. [45]

    Meng, W. (2025). weiyaomeng/gpt-radar-app: Public release v1.1 (update)

  38. [46]

    Metzger, M. J. (2007). Making sense of credibility on the web: Models for evaluating online information and recommendations for future research.Journal of the American society for information science and technology, 58(13):2078–2091

  39. [47]

    Metzger, M. J. and Flanagin, A. J. (2015). Psychological approaches to credibility assessment online.The handbook of the psychology of communication technology, pages 445–466

  40. [48]

    Moser, C., Den Hond, F., and Lindebaum, D. (2022). Morality in the age of artificially intelligent algorithms.Academy of Management Learning & Education, 21(1):139–155

  41. [49]

    Mudambi, S. M. and Schuff, D. (2010). Research note: What makes a helpful online review? a study of customer reviews on amazon. com.MIS quarterly, pages 185–200

  42. [50]

    Newman, G. E. (2019). The psychology of authenticity.Review of General Psychology, 23(1):8–18

  43. [51]

    Nica-Avram, G., Ljevar, V ., Harvey, J., Branco-Illodo, I., Gallage, S., and Goulding, J. (2022). Ill-fated interactions: modeling complaints on a food waste fighting platform. In2022 IEEE International Conference on Big Data (Big Data), pages 4105–4109. IEEE. O’Reilly, K. and...

  44. [52]

    A., Yau, A., Lavertu, L., Stylos, N., Oliver, S., Angell, R., de Regt, A., Gao, L., et al

    Osadchaya, E., Marder, B., Yule, J. A., Yau, A., Lavertu, L., Stylos, N., Oliver, S., Angell, R., de Regt, A., Gao, L., et al. (2024). To chatgpt, or not to chatgpt: Navigating the paradoxes of generative ai in the advertising industry.Business Horizons, 67(5):571–581

  45. [53]

    Packard, V . (1957). The hidden persuaders.David McKay Co

  46. [54]

    M., Laroche, M., and Braune, E

    Sahut, J. M., Laroche, M., and Braune, E. (2024). Antecedents and consequences of fake reviews in a marketing approach: An overview and synthesis

  47. [55]

    M., Jung, S.-g., and Jansen, B

    Salminen, J., Kandpal, C., Kamel, A. M., Jung, S.-g., and Jansen, B. J. (2022). Creating and detecting fake reviews of online products.Journal of Retailing and Consumer Services, 64:102771

  48. [56]

    Schneiders, E., Seabrooke, T., Krook, J., Hyde, R., Leesakul, N., Clos, J., and Fischer, J. (2024). Objection overruled! lay people can distinguish large language models from lawyers, but still favour advice from an llm.arXiv preprint arXiv:2409.07871

  49. [57]

    Shukla, A. D. and Goh, J. M. (2024). Fighting fake reviews: Authenticated anonymous reviews using identity verification.Business Horizons, 67(1):71–81

  50. [58]

    J., Deveci, C

    Sikander, B., Baker, J. J., Deveci, C. D., Lund, L., and Rosenberg, J. (2023). Chatgpt-4 and human researchers are equal in writing scientific introduction sections: a blinded, randomized, non-inferiority controlled study.Cureus, 15(11). 20

  51. [59]

    (2025).Consumer

    Smith, A., Lukinova, E., Harvey, J., Smith, G., Mansilla, R., Goulding, J., and Nica-Avram, G. (2025).Consumer

  52. [60]

    (2018).Life 3.0: Being human in the age of artificial intelligence

    Tegmark, M. (2018).Life 3.0: Being human in the age of artificial intelligence. Vintage

  53. [61]

    P., Frisch, P., Domhan, T., and Federico, M

    Thompson, B., Dhaliwal, M. P., Frisch, P., Domhan, T., and Federico, M. (2024). A shocking amount of the web is machine translated: Insights from multi-way parallelism.arXiv preprint arXiv:2401.05749

  54. [62]

    Trott, S., Jones, C., Chang, T., Michaelov, J., and Bergen, B. (2023). Do large language models know what humans know?Cognitive Science, 47(7):e13309. UK Department for Business and Trade (2023). Investigating the prevalence and impact of fake reviews. Accessed: 2023-04-25

  55. [63]

    Ungoed-Thomas, J. (2025). Uk bans £2.2bn ‘sneaky’ fees and fake reviews for online products.The Guardian. Accessed: 2025-04-06

  56. [64]

    Vernuccio, M., Patrizi, M., and Pastore, A. (2023). Delving into brand anthropomorphisation strategies in the experiential context of name-brand voice assistants.Journal of Consumer Behaviour, 22(5):1074–1083

  57. [65]

    S., and Wong, D

    Wu, J., Yang, S., Zhan, R., Yuan, Y ., Chao, L. S., and Wong, D. F. (2025). A survey on llm-generated text detection: Necessity, methods, and future directions.Computational Linguistics, pages 1–66

  58. [66]

    F., Xanthopoulos, P., Karampelas, P., and Bakamitsos, G

    Xylogiannopoulos, K. F., Xanthopoulos, P., Karampelas, P., and Bakamitsos, G. A. (2024). Chatgpt paraphrased product reviews can confuse consumers and undermine their trust in genuine reviews. can you tell the difference? Information Processing & Management, 61(6):103842

  59. [67]

    and Zhang, X

    Zhu, F. and Zhang, X. (2010). Impact of online consumer reviews on sales: The moderating role of product and consumer characteristics.Journal of marketing, 74(2):133–148. 21

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.