Pith. sign in

REVIEW 3 major objections 5 minor 36 references

How Large Language Models play humans in online conversations: a simulated study of the 2016 US politics on Reddit

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read GPT-4 impersonates Reddit users convincingly but generates consensus more readily than dissent.

desk verdict Solid, useful simulation study on Reddit that overclaims its main finding because the labels come from the same model that wrote the comments. read the letter →

arxiv 2506.21620 v1 pith:II4F6LUF submitted 2025-06-23 cs.CL cs.AIcs.CYcs.SIphysics.soc-ph

Daniele Cirulli , Giulio Cimini , Giovanni Palermo This is my paper
classification cs.CLcs.AIcs.CYcs.SIphysics.soc-ph
keywords largelanguagemodelsGPT-4Redditpoliticalalignmentconsensusbiasembeddingspacebotdetection2016USelection
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

This paper asks whether a large language model can pass as a human participant in politically charged online discussions, and what traces it leaves if it can. Using Reddit threads from the 2016 US presidential election, the authors prompt GPT-4 to write comments as a real user, as a pro-candidate supporter, or as an anti-candidate opponent, and compare the output to the original human comments. The central result is an asymmetry: GPT-4 reproduces the style of human discourse, but it generates consensus much more readily than dissent, so even an explicitly anti-candidate identity yields mostly neutral or pro-candidate text. The paper also finds that real and synthetic comments separate into distinct clusters in semantic embedding space, offering a quantitative route toward catching such bots, even though the generated texts are hard to tell apart by eye.

What carries the argument

The experimental machinery combines three impersonation scenarios that differ only in the user history fed to the prompt (real past comments, a fictitious pro-candidate history, or a fictitious anti-candidate history), plus a null model with no history. The generated texts are labeled by GPT-4 itself with three three-valued tags (party alignment, sentiment, violence) using in-context learning, and are embedded with a text-embedding model, projected with a dimensionality-reduction technique, and finally separated with a linear support vector machine in a five-class task. The load-bearing object is the prompt-to-history pairing: it isolates the effect of the supplied political identity on the model's output.

What would settle it

Have a panel of human annotators label the same real and generated comments for candidate alignment, and compare those labels with GPT-4's; if human labels raise the anti-candidate share of Scenario 3 toward the pro-candidate share of Scenario 2, or if the null model stops looking neutral, the paper's consensus-over-dissent claim collapses. A second check: train the five-class embedding classifier on one subreddit and test on the other; if accuracy drops to chance, the reported separability may be dataset-specific rather than a general bot trace.

Watch

Extended reading notes

Core claim

GPT-4, when asked to impersonate a user in a Reddit thread, produces comments that look human but are measurably different from what people actually write. The most important difference is that the model avoids dissent: with a fictitious history that strongly opposes the subreddit's candidate, the generated comment is still anti-candidate only about 32% of the time on the Clinton side and 22% on the Trump side, while a supportive history produces pro-candidate comments over half the time. With a real user's history, generated comments become more pro-candidate than the user's actual comments, even for users whose history is anti-candidate. At the same time, synthetic comments occupy their own region of the embedding space: a linear classifier trained on text embeddings separates real comments from four types of generated ones with roughly 79% average accuracy, and a two-dimensional projection shows the clusters clearly, though human inspection cannot reliably distinguish the texts.

Load-bearing premise

The entire consensus-versus-dissent result rests on GPT-4's own three-number labels for party alignment, sentiment, and violence, which are used without validation against human annotators; if the evaluator systematically calls its own generated comments neutral or pro-candidate, the asymmetry would be an artifact of the measurer.

Editorial extensions

If this is right

  • A partisan bot built on GPT-4 could seed supportive comments in a friendly community more reliably than it could sow dissent in an enemy community.
  • Because generated comments cluster separately from real ones in embedding space, a linear classifier offers a practical, though not perfect, screen for LLM-written political posts.
  • The near-total absence of violent generated text suggests built-in content constraints carry over even when the model is explicitly role-playing an aggressive user.
  • The finding that longer prompts push generated comments toward the subreddit's leaning implies that context-rich threads make consensus easier to elicit.

Reading between the lines

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

  • The consensus bias the authors observe may reflect the model's alignment training rather than a property of language itself; a useful next test would be to run the same prompts on models with different reinforcement-learning-from-human-feedback policies and compare Scenario 3 dissent rates.
  • The embedding-space separation could be a moving target: as future models are trained to imitate human style more closely, the linear classifier accuracy may degrade, so the method should be re-benchmarked periodically.
  • If the paper's result generalizes, platform moderators could treat unusually high dissent rates in a community as a signal of human activity, while unusually consensus-bound responding might flag automation.
  • The five-class classifier was trained and tested on the same subreddits; a fairer estimate of detection value would require a held-out election cycle or a different platform.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper evaluates GPT-4's ability to generate Reddit comments in the context of the 2016 US presidential election, using three scenarios: impersonation of real users (Scenario 1), supportive bots (Scenario 2), and dissenting bots (Scenario 3), plus a null model with no user history. The authors classify generated and real comments for party alignment, sentiment, and violence using GPT-4 itself, and compare their semantic embeddings via t-SNE and a linear SVC. The central claims are that GPT-4 produces realistic comments but tends to create consensus more easily than dissent, and that real and artificial comments are separable in embedding space despite being indistinguishable by human inspection.

Significance. If the claims hold, the paper offers a useful empirical benchmark for LLM behavior in politically loaded online settings, with implications for bot detection and discourse manipulation. The use of real Reddit data, a null model, and multiple prompting conditions is a strength, and the supplementary SVC analysis provides a quantitative complement to the t-SNE visualization. However, the main finding is only as strong as the validity of GPT-4's self-ratings, which are not validated against human labels; a human-annotation study and statistical error bars would make the contribution far more credible.

major comments (3)
  1. [Section III-A (Party Alignment classification prompt)] The party-alignment and sentiment labels used throughout the paper are produced by GPT-4 with a single in-context example that is explicitly abusive ('Hillary Clinton is a whore!' -> -1, -1, 1). This anchoring is likely to teach the model to reserve the 'anti-candidate' category for profanity-laced attacks and to classify measured or implicit criticism as neutral, which would inflate the neutral share and deflate the anti-candidate share. The paper reports no human validation, inter-annotator agreement, or error analysis for these labels. Since the 'consensus over dissent' finding rests entirely on these labels, the authors must provide human-annotated validation (e.g., a few hundred comments labeled by 2-3 annotators) and report agreement metrics and a confusion matrix for the GPT-4 classifier.
  2. [Figures 2 and 3, Section III-A] All percentages in Figures 2 and 3 are averages over 5 runs with no error bars, confidence intervals, or significance tests. The text draws strong conclusions from these point estimates: 'generated dissenting comments almost vanish' (Scenario 1: 0% and 1% anti-candidate) and 'the anti-candidate counterpart is steadily below 30%' (Scenario 3). With only five runs, these differences could be within run-to-run variability; the authors should report the per-run spread (e.g., standard deviation or bootstrap intervals) and, ideally, a permutation test comparing the generated anti-candidate shares against the real-comment baseline.
  3. [Section III-B] The claim that real and generated comments are 'indistinguishable by manual inspection' is not supported by a systematic human study. The text states only that the authors themselves found it difficult to tell them apart; no protocol, sample size, or accuracy measure is given. Additionally, the 'clear separation' in the t-SNE plots (Figure 4) is a visual judgement. The quantitative SVC experiment (Supplementary S8) is more convincing, but it is not cited in the main text and does not appear in the main figures. The authors should integrate a quantitative evaluation of separability (e.g., the SVC accuracy with confidence intervals) into the main text and either run or explicitly retract the anonymous human-inspection claim.
minor comments (5)
  1. [Section II-B] The paper does not clearly state how many target comments were simulated per user; it reports that 100 and 387 users contributed 5,220 and 5,488 comments in 2016, but the number of simulated outputs per run is never specified. This makes it difficult to assess the effective sample sizes behind the percentages in Figures 2 and 3.
  2. [Section III-A.1] The sentence 'The model generates a little more dissent in Trump’s Subreddit (1.8% against 0.6% for Clinton), yet this difference may be explained by looking at the classification of the posts' presents a post-hoc explanation without testing; a simple bootstrap comparison would be more appropriate.
  3. [Section III-B] The t-SNE figures (Figure 4) do not report the perplexity or other hyperparameters, and the axis scales are not labeled with counts; for reproducibility, these details should be given.
  4. [Conclusion] The statement 'this behavior holds for both Trump and Clinton’s Subreddits' is too strong given the visible differences between the two subreddits in Scenario 3 (32% vs 22% anti-candidate); the conclusion should be qualified.
  5. [General] No code or data are released (arXiv submission, no links), which prevents external reproduction of the t-SNE and SVC analyses; a reproducibility statement would be desirable.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's claims rest on direct measurements, not on fitted parameters or self-citations.

full rationale

The paper's central claim—that GPT-4 tends to create consensus more easily than dissent—is an empirical measurement: comments are generated under different prompting scenarios and then classified for party alignment, sentiment, and violence using a GPT-4 prompt (Section III-A). This is a single-model evaluation loop, since the same model family generates and classifies the comments, and the classification prompt's in-context example ('Hillary Clinton is a whore!' mapped to -1, -1, 1) may bias the model to reserve the anti-candidate label for overtly abusive language. However, this is a validity threat, not circularity: the classification labels are not defined in terms of the conclusion, no parameter is fitted to the target result, and the reported percentages are direct experimental outputs rather than quantities reconstructed from the inputs. The embedding-space separation and the linear SVC five-class accuracy (about 79%) are obtained from real versus generated comments using a separate embedding model and standard supervised training, with real human comments serving as an external benchmark; these results do not reduce to the prompting inputs. The manual indistinguishability observation in Section III-B is anecdotal and not systematically validated, but it is presented as informal observation rather than as a derived prediction. No load-bearing step is justified by a self-citation, and no equation or fitted value reappears as the conclusion. Therefore, no circular step can be exhibited with the required specificity, and the appropriate finding is no significant circularity (score 0), with the unvalidated classification instrument noted as a methodological risk rather than a circularity.

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

This is an empirical study; there are no fitted numeric parameters. The main assumptions concern the validity of GPT-4 as a classifier, the representativeness of the selected users, and the semantics captured by the embedding model.

assumptions (4)
  • domain assumption GPT-4's classifications of party alignment, sentiment, and violence are accurate for both real and generated comments.
    Section III-A introduces the classification prompt but provides no validation against human labels or a gold standard; the main findings depend on these labels.
  • domain assumption The selected users (100 for Clinton, 387 for Trump) and their 2015 comments are a representative behavioral template for 2016 comments.
    Section II-A defines the user selection; the imbalance between subreddits and the reliance on a single year of history are not discussed.
  • domain assumption OpenAI text-embedding-3-small embeddings capture semantically meaningful differences relevant to authorship.
    Section III-B uses this embedding to claim separation; while the SVC supports it, the embedding's invariance properties are not analyzed.
  • domain assumption The null model (no user history) is an appropriate baseline for isolating the effect of user-history prompting.
    Section II-C describes the null scenario; the authors do not test alternative baselines.

how reviews work

0 comments
Cite this review

Pith. "Pith review of How Large Language Models play humans in online conversations: a simulated study of the 2016 US politics on Reddit." pith.science (2026). https://pith.science/paper/II4F6LUF

@misc{pith2026250621620,
  author       = {Pith},
  title        = {Pith review of: How Large Language Models play humans in online conversations: a simulated study of the 2016 US politics on Reddit},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/II4F6LUF}},
  note         = {Machine review of arXiv:2506.21620}
}
read the original abstract

Large Language Models (LLMs) have recently emerged as powerful tools for natural language generation, with applications spanning from content creation to social simulations. Their ability to mimic human interactions raises both opportunities and concerns, particularly in the context of politically relevant online discussions. In this study, we evaluate the performance of LLMs in replicating user-generated content within a real-world, divisive scenario: Reddit conversations during the 2016 US Presidential election. In particular, we conduct three different experiments, asking GPT-4 to generate comments by impersonating either real or artificial partisan users. We analyze the generated comments in terms of political alignment, sentiment, and linguistic features, comparing them against real user contributions and benchmarking against a null model. We find that GPT-4 is able to produce realistic comments, both in favor of or against the candidate supported by the community, yet tending to create consensus more easily than dissent. In addition we show that real and artificial comments are well separated in a semantically embedded space, although they are indistinguishable by manual inspection. Our findings provide insights on the potential use of LLMs to sneak into online discussions, influence political debate and shape political narratives, bearing broader implications of AI-driven discourse manipulation.

Figures

Figures reproduced from arXiv: 2506.21620 by the authors.

Figure 1
Figure 1. Illustration of a comment simulations in a conversation using multiple prompt types. We simulate a target user’s response [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Party alignment classification of generated comments across different scenarios and conditions, averaged over 5 runs. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Sentiment classification of generated comments across different scenarios and conditions, averaged over 5 runs. The [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: t-SNE projection of real and generated comments in [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

36 extracted references · 18 canonical work pages

  1. [1]

    Lewis, James M

    Yongjun Xu, Xin Liu, Xin Cao, Changping Huang, Enke Liu, Sen Qian, Xingchen Liu, Yanjun Wu, Fengliang Dong, Cheng-Wei Qiu, Junjun Qiu, Keqin Hua, Wentao Su, Jian Wu, Huiyu Xu, Yong Han, Chenguang Fu, Zhigang Yin, Miao Liu, Ronald Roepman, Sabine Dietmann, Marko Virta, Fredrick Kengara, Ze Zhang, Lifu Zhang, Taolan Zhao, Ji Dai, Jialiang Yang, Liang Lan, M...

    work page 2021

  2. [2]

    Willcocks

    Sam Bond-Taylor, Adam Leach, Yang Long, and Chris G. Willcocks. Deep generative modelling: A comparative review of vaes, gans, normal- izing flows, energy-based and autoregressive models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(11):7327–7347, 2022

    work page 2022

  3. [3]

    A survey of large language models

    Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, Xiaolei Wang, Yupeng Hou, Yingqian Min, Beichen Zhang, Junjie Zhang, Zican Dong, Yifan Du, Chen Yang, Yushuo Chen, Zhipeng Chen, Jinhao Jiang, Ruiyang Ren, Yifan Li, Xinyu Tang, Zikang Liu, Peiyu Liu, Jian-Yun Nie, and Ji-Rong Wen. A survey of large language models. arXiv preprint arXiv:2303.18223, 2023

    arXiv 2023

  4. [4]

    Gpt-4 technical report

    OpenAI et al. Gpt-4 technical report. arXiv preprint arXiv:2303.08774, 2024

    arXiv 2024

  5. [5]

    Llama: Open and efficient foundation language models

    Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie- Anne Lachaux, Timoth´ee Lacroix, Baptiste Rozi `ere, Naman Goyal, Eric Hambro, Faisal Azhar, Aurelien Rodriguez, Armand Joulin, Edouard Grave, and Guillaume Lample. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 , 2023

    arXiv 2023

  6. [6]

    Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, L ´elio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timoth ´ee Lacroix, and William El Sayed. Mistral 7B. arXiv preprint arX...

    arXiv

  7. [7]

    Gemini: A family of highly capable multimodal models

    Gemini Team. Gemini: A family of highly capable multimodal models. arXiv preprint arXiv:2312.11805 , 2023

    arXiv 2023

  8. [8]

    Large language models in complex system design

    Alejandro Pradas Gomez, Petter Krus, Massimo Panarotto, and Ola Isaksson. Large language models in complex system design. Proceed- ings of the Design Society , 4:2197–2206, 2024

    work page 2024

Show all 36 references
  1. [9]

    LLMs and generative agent-based models for complex systems research

    Yikang Lu, Alberto Aleta, Chunpeng Du, Lei Shi, and Yamir Moreno. LLMs and generative agent-based models for complex systems research. Physics of Life Reviews , 51:283–293, 2024

    2024

  2. [10]

    Large language models on graphs: A comprehensive survey

    Bowen Jin, Gang Liu, Chi Han, Meng Jiang, Heng Ji, and Jiawei Han. Large language models on graphs: A comprehensive survey. IEEE Transactions on Knowledge and Data Engineering , 36(12):8622–8642, 2024

    2024

  3. [11]

    Emergence of scale-free networks in social interactions among large language models

    Giordano De Marzo, Luciano Pietronero, and David Garcia. Emergence of scale-free networks in social interactions among large language models. arXiv preprint arXiv:2312.06619 , 2023

    2023 arXiv

  4. [12]

    Emergence of scaling in random networks

    Albert-L ´aszl´o Barab ´asi and R ´eka Albert. Emergence of scaling in random networks. Science, 286(5439):509–512, 1999

    1999

  5. [13]

    Networks

    Mark Newman. Networks. Oxford University Press, 2010

    2010

  6. [14]

    Statistical physics of social dynamics

    Claudio Castellano, Santo Fortunato, and Vittorio Loreto. Statistical physics of social dynamics. Reviews of Modern Physics, 81(2):591–646, 2009

    2009

  7. [15]

    Social simulacra: Creating populated prototypes for social computing systems

    Joon Sung Park, Lindsay Popowski, Carrie Cai, Meredith Ringel Morris, Percy Liang, and Michael S Bernstein. Social simulacra: Creating populated prototypes for social computing systems. In Proceedings of the 35th Annual ACM Symposium on User Interface Software and Technology, ...

    2022

  8. [16]

    The dynamics of social conventions in llm populations: Spontaneous emergence, collective biases and tipping points

    Ariel Flint Ashery, Luca Maria Aiello, and Andrea Baronchelli. The dynamics of social conventions in llm populations: Spontaneous emergence, collective biases and tipping points. arXiv preprint arXiv:2410.08948, 2024

    2024 arXiv

  9. [17]

    Spontaneous emergence of agent individuality through social interactions in llm-based communities

    Ryosuke Takata, Atsushi Masumori, and Takashi Ikegami. Spontaneous emergence of agent individuality through social interactions in llm-based communities. arXiv preprint arXiv:2411.03252 , 2024

    2024 arXiv

  10. [18]

    Nicer than humans: How do large language models behave in the prisoner’s dilemma? arXiv preprint arXiv:2406.13605 , 2024

    Nicol ´o Fontana, Francesco Pierri, and Luca Maria Aiello. Nicer than humans: How do large language models behave in the prisoner’s dilemma? arXiv preprint arXiv:2406.13605 , 2024

    2024 arXiv

  11. [19]

    Gtbench: Uncovering the strategic reasoning limitations of llms via game-theoretic evaluations

    Jinhao Duan, Renming Zhang, James Diffenderfer, Bhavya Kailkhura, Lichao Sun, Elias Stengel-Eskin, Mohit Bansal, Tianlong Chen, and Kaidi Xu. Gtbench: Uncovering the strategic reasoning limitations of llms via game-theoretic evaluations. arXiv preprint arXiv:2402.12348 , 2024

    2024 arXiv

  12. [20]

    Game-theoretic llm: Agent workflow for negotiation games

    Wenyue Hua, Ollie Liu, Lingyao Li, Alfonso Amayuelas, Julie Chen, Lucas Jiang, Mingyu Jin, Lizhou Fan, Fei Sun, William Wang, et al. Game-theoretic llm: Agent workflow for negotiation games. arXiv preprint arXiv:2411.05990, 2024

    2024 arXiv

  13. [21]

    Alympics: Language agents meet game theory

    Shaoguang Mao, Yuzhe Cai, Yan Xia, Wenshan Wu, Xun Wang, Fengyi Wang, Tao Ge, and Furu Wei. Alympics: Language agents meet game theory. arXiv preprint arXiv:2311.03220 , 2023

    2023 arXiv

  14. [22]

    Towards the terminator economy: Assessing job exposure to ai through llms

    Emilio Colombo, Fabio Mercorio, Mario Mezzanzanica, and Antonio Serino. Towards the terminator economy: Assessing job exposure to ai through llms. arXiv preprint arXiv:2407.19204 , 2024

    2024

  15. [23]

    Generative AI and jobs: a global analysis of potential effects on job quantity and quality

    Pawel Gmyrek, Janine Berg, and David Bescond. Generative AI and jobs: a global analysis of potential effects on job quantity and quality . ILO working paper 96, 2023

    2023

  16. [24]

    Follow the money: a startup-based measure of ai exposure across occupations, industries and regions

    Enrico Maria Fenoaltea, Dario Mazzilli, Aurelio Patelli, Angelica Sbardella, Andrea Tacchella, Andrea Zaccaria, Marco Trombetti, and Luciano Pietronero. Follow the money: a startup-based measure of ai exposure across occupations, industries and regions. arXiv preprint arXiv:24...

    2024 arXiv

  17. [25]

    A complete survey on llm-based ai chatbots

    Sumit Kumar Dam, Choong Seon Hong, Yu Qiao, and Chaoning Zhang. A complete survey on llm-based ai chatbots. arXiv preprint arXiv:2406.16937, 2024

    2024 arXiv

  18. [26]

    Gonzalez, and Ion Stoica

    Wei-Lin Chiang, Lianmin Zheng, Ying Sheng, Anastasios Nikolas An- gelopoulos, Tianle Li, Dacheng Li, Hao Zhang, Banghua Zhu, Michael Jordan, Joseph E. Gonzalez, and Ion Stoica. Chatbot arena: An open platform for evaluating llms by human preference. arXiv preprint arXiv:2403.0...

    2024 arXiv

  19. [27]

    Pierrehumbert

    Valentin Hofmann, Hinrich Sch ¨utze, and Janet B. Pierrehumbert. The reddit politosphere: A large-scale text and network resource of online political discourse. Proceedings of the International AAAI Conference on Web and Social Media , 16:1259–1267, 2022

    2022

  20. [28]

    The pushshift reddit dataset

    Jason Baumgartner, Savvas Zannettou, Brian Keegan, Megan Squire, and Jeremy Blackburn. The pushshift reddit dataset. arXiv preprint arXiv:2001.08435, 2020

    2001 arXiv

  21. [29]

    Evaluating GPT-3 generated explanations for hateful content moderation

    Han Wang, Ming Shan Hee, Md Rabiul Awal, Kenny Tsu Wei Choo, and Roy Ka-Wei Lee. Evaluating GPT-3 generated explanations for hateful content moderation. In Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence , IJCAI-2023, page 6255–6263, 2023

    2023

  22. [30]

    Visualizing data using t-SNE

    Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-SNE. Journal of machine learning research , 9(11), 2008

    2008

  23. [31]

    ChatGPT: a tool for scientific writing or a threat to integrity? The British Journal of Radiology , 96(1152), 2023

    Tha ´ısa Pinheiro Silva, Tha ´ıs S C Ocampo, Caio Alencar-Palha, Chris- tiano Oliveira-Santos, Wilton Mitsunari Takeshita, and Matheus L Oliveira. ChatGPT: a tool for scientific writing or a threat to integrity? The British Journal of Radiology , 96(1152), 2023

    2023

  24. [32]

    More human than human: measuring chatgpt political bias

    Fabio Motoki, Valdemar Pinho Neto, and Victor Rodrigues. More human than human: measuring chatgpt political bias. Public Choice, 198(1):3– 23, 2024

    2024

  25. [33]

    What does the bot say? opportunities and risks of large language models in social media bot detection

    Shangbin Feng, Herun Wan, Ningnan Wang, Zhaoxuan Tan, Minnan Luo, and Yulia Tsvetkov. What does the bot say? opportunities and risks of large language models in social media bot detection. arXiv preprint arXiv:2402.00371, 2024

    2024 arXiv

  26. [34]

    Christakis, David Garcia, Amit Goldenberg, Yara Kyrychenko, Kevin Leyton-Brown, Nina Lutz, Gary Marcus, Filippo Menczer, Gordon Pennycook, David G

    Daniel Thilo Schroeder, Meeyoung Cha, Andrea Baronchelli, Nick Bostrom, Nicholas A. Christakis, David Garcia, Amit Goldenberg, Yara Kyrychenko, Kevin Leyton-Brown, Nina Lutz, Gary Marcus, Filippo Menczer, Gordon Pennycook, David G. Rand, Frank Schweitzer, Christopher Summerfie...

    2025

  27. [35]

    in reply to:

    Nils Reimers and Iryna Gurevych. Sentence-bert: Sentence embeddings using siamese bert-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing , pages 3982–3992. Association for Computational Linguistics, Nov 2019. Supplementary Info...

    1997

  28. [36]

    and shows that clusters of real and generated text for different temperatures, remain distinct even when employing the less expressive SBERT model rather than the OpenAI encoder, thereby underscoring the robustness of our classification. 5 × 10 1 6 × 10 1 7 × 10 1 8 × 10 1 Per...

    1947

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.