The Political Biases of ChatGPT
Open accessThe DOI matches its Crossref record; 12 of the 17 references it cites were found in a registry or an index, and 5 are not checked yet.
Abstract
Recent advancements in Large Language Models (LLMs) suggest imminent commercial applications of such AI systems where they will serve as gateways to interact with technology and the accumulated body of human knowledge. The possibility of political biases embedded in these models raises concerns about their potential misusage. In this work, we report the results of administering 15 different political orientation tests (14 in English, 1 in Spanish) to a state-of-the-art Large Language Model, the popular ChatGPT from OpenAI. The results are consistent across tests; 14 of the 15 instruments diagnose ChatGPT answers to their questions as manifesting a preference for left-leaning viewpoints. When asked explicitly about its political preferences, ChatGPT often claims to hold no political opinions and to just strive to provide factual and neutral information. It is desirable that public facing artificial intelligence systems provide accurate and factual information about empirically verifiable issues, but such systems should strive for political neutrality on largely normative questions for which there is no straightforward way to empirically validate a viewpoint. Thus, ethical AI systems should present users with balanced arguments on the issue at hand and avoid claiming neutrality while displaying clear signs of political bias in their content.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Available data point to ChatGPT output expressing more progressive than traditional political views [38][39][40].”
From The Impact of Artificial Intelligence on Human Sexuality: A Five-Year Literature Review 2020–2024 (2024) · cites this paper as [38]
The cited source says what the claim attributes to it.
In the citing paper
“It shared findings from a study where 15 different political orientation tests were administered to ChatGPT. The results were significant, with 14 out of the 15 tests indicating that ChatGPT's responses inclined towards left-leaning viewpoints.”
From Generative AI in Higher Education: The ChatGPT Effect (2024) · cites this paper as Rozado (2023)
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404 · cited 1×
“In particular, Large Language Models (LLMs) (Zhou et al. 2022) based on the Transformer architecture (Vaswani et al. 2017) have pushed the state-of-the-art substantially in natural language tasks such as machine translation (Dabre et al. 2020), sentiment analysis (Ain et al. 2017), name entity…”
BackgroundA Survey on Deep Learning for Named Entity Recognition
Li, J., Sun, A., Han, J., & Li, C. (2022) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2020.2981314 · Cited by 1,567
Chatbots: History, technology, and applications
Adamopoulou, E., & Moussiades, L. (2020) · Machine Learning with Applications · doi:10.1016/j.mlwa.2020.100006 · Cited by 1,090 · cited 1×
“…architecture (Vaswani et al. 2017) have pushed the state-of-the-art substantially in natural language tasks such as machine translation (Dabre et al. 2020), sentiment analysis (Ain et al. 2017), name entity recognition (Li et al. 2022), or dialogue bots (Adamopoulou and Moussiades 2020).”
BackgroundDeep Learning vs. Traditional Computer Vision
O’Mahony, N., Campbell, S., Carvalho, A., Harapanahalli, S., Hernandez, G. V., Krpalkova, L., Riordan, D., & Walsh, J. (2020) · Advances in Intelligent Systems and Computing · doi:10.1007/978-3-030-17795-9_10 · Cited by 1,025
Sentiment Analysis Using Deep Learning Techniques: A Review
Tul, Q., Ali, M., Riaz, A., Noureen, A., Kamranz, M., Hayat, B., & Rehman, A. (2017) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2017.080657 · Cited by 334 · cited 1×
“…architecture (Vaswani et al. 2017) have pushed the state-of-the-art substantially in natural language tasks such as machine translation (Dabre et al. 2020), sentiment analysis (Ain et al. 2017), name entity recognition (Li et al. 2022), or dialogue bots (Adamopoulou and Moussiades 2020).”
BackgroundAlgorithmic Bias: From Discrimination Discovery to Fairness-aware Data Mining
Hajian, S., Bonchi, F., & Castillo, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2945386 · Cited by 334 · cited 1×
“The topic of algorithmic bias has received an increasing amount of attention in the machine learning academic literature ( Wikipedia 2023a; Kirkpatrick 2016; Cowgill and Tucker 2017; Garcia 2016; Hajian et al. 2016).”
BackgroundLarge Language Models Are Human-Level Prompt Engineers
Zhou, Y., Muresanu, A. I., Han, Z., Paster, K., Pitis, S., Chan, H., & Ba, J. (2022) · arXiv · doi:10.48550/arxiv.2211.01910 · Cited by 330 · cited 1×
“In particular, Large Language Models (LLMs) (Zhou et al. 2022) based on the Transformer architecture (Vaswani et al. 2017) have pushed the state-of-the-art substantially in natural language tasks such as machine translation (Dabre et al. 2020), sentiment analysis (Ain et al. 2017), name entity…”
BackgroundA Survey of Multilingual Neural Machine Translation
Dabre, R., Chu, C., & Kunchukuttan, A. (2021) · ACM Computing Surveys · doi:10.1145/3406095 · Cited by 294 · cited 2×
“The 2012–2022 decade has witnessed spectacular improvements in AI, from computer vision (O’Mahony et al. 2020), to machine translation (Dabre et al. 2020), to generative models for images ( Wikipedia 2023c) and text ( Wikipedia 2022).”
BackgroundGarcia, M. (2016) · World Policy Journal · doi:10.1215/07402775-3813015 · Cited by 257 · cited 1×
“The topic of algorithmic bias has received an increasing amount of attention in the machine learning academic literature ( Wikipedia 2023a; Kirkpatrick 2016; Cowgill and Tucker 2017; Garcia 2016; Hajian et al. 2016).”
BackgroundThe American Journalist in the Digital Age: Another Look at U.S. News People
Weaver, D. H., Willnat, L., & Wilhoit, G. C. (2019) · Journalism & Mass Communication Quarterly · doi:10.1177/1077699018778242 · Cited by 174 · cited 1×
“…of professionals working in these institutions are politically left-leaning (Reuters Institute for the Study of Journalism n.d.; Hopmann et al. 2010; Weaver et al. 2019; Langbert 2018; Archive et al. 2021; Schoffstall 2022; American Enterprise Institute—AEI (blog) n.d.; The Harvard Crimson n.d.).”
Background
Research works printed but not linked: 2
- Bender, Emily M., Gebru, Timnit, McMillan-Major, Angelina, and Shmitchell, Shmargaret (2021). FAccT ’21: Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, Association for Computing Machinery.
- Hajian, Sara, Bonchi, Francesco, and Castillo, Carlos (2016). KDD ’16: Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM.
Cited by
Items 101–200 of 282
Abbas, A., Azar, B. B., Mahrishi, M., Núñez, J. L. M., & Mishra, D. (2025) · Equilibrium Quarterly Journal of Economics and Economic Policy · doi:10.24136/eq.3551 · Cited by 21
“…the addressing of algorithmic bias and fairness, the formation of collaborations and partnerships with stakeholders, the integration of AI governance into the curriculum, and concern about political prejudice and discrimination (Kumi-Yeboah et al. , 2023; Atenas et al. , 2023; Rozado, 2023).”
Kostka, G., & Zhou, H. (2026) · Technology in Society · doi:10.1016/j.techsoc.2026.103379 · Cited by 18
Kouzelis, L. R., & Spantidi, O. (2024) · Applied Sciences · doi:10.3390/app14156405 · Cited by 12
Munnik, J. B., & Noorbhai, H. (2024) · South African Journal of Psychology · doi:10.1177/00812463231223427 · Cited by 10
“There is also a growing concern about political biases influencing AI algorithms, as demonstrated by Peters (2022) and Rozado (2023).”
BackgroundControl search rankings, control the world: what is a good search engine?
Coghlan, S., Chia, H. X., Scholer, F., & Spina, D. (2025) · AI and Ethics · doi:10.1007/s43681-025-00695-8 · Cited by 10
“Further, LLM-based SEs must be very careful to avoid harmful biased outputs [76, 83 ].”
Haxvig, H. A., D’Andrea, V., & Teli, M. (2025) · International Journal of Human-Computer Studies · doi:10.1016/j.ijhcs.2025.103651 · Cited by 14
Security and Privacy of Large Language Models: Threat Taxonomy, Ethical Implications, and Governance
Pribisalić, M., & Martinčić-Ipšić, S. (2026) · AI · doi:10.3390/ai7050152 · Cited by 9
Is ChatGPT a friend or foe in the war on misinformation?: A South African perspective
Senekal, B., & Brokensha, S. (2023) · Communicare: Journal for Communication Studies in Africa · doi:10.36615/jcsa.v42i2.2437 · Cited by 7
“One significant area of concern relates to whether ChatGPT or other LLMs could in future be used as main sources of information, with the former being proposed as a source of information in competition with search engines (Borji, 2023; Rozado, 2023).”
BackgroundResultInfluentialSpennemann, D. H. R. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5393987 · Cited by 9
Ethics and the use of generative AI in professional editing
Otmar, R., Michael, R., Mullins, S., & Day, K. (2024) · AI and Ethics · doi:10.1007/s43681-024-00521-7 · Cited by 11
“Other studies have reported undue influence through word embeddings and algorithms in older and newer versions of LLMs, including gender bias [49], racial bias [21] and political bias [55].”
Development of Linguistic Competence in English for Specific Purposes Through ChatGPT: A Case Study
Enamorado, M. Á. J. (2025) · Journal of language and Education · doi:10.17323/jle.2025.23745 · Cited by 21
“Howev - er, this capability of AI has raised concern about the possi - bility of its misuse for creating false news or manipulating public opinion (Floridi & Chiriati, 2020) and for its political bias (Rozado, 2023).”
Can AI-Attributed News Challenge Partisan News Selection? Evidence from a Conjoint Experiment
Zoizner, A., Matthes, J., Corbu, N., de Vreese, C. H., Esser, F., Koç-Michalska, K., Schemer, C., Theocharis, Y., & Zilinsky, J. (2025) · The International Journal of Press/Politics · doi:10.1177/19401612251342679 · Cited by 11
“AI outputs have been shown to reflect gender and racial biases, and often present politically left-leaning orientations (e.g., Abid et al. 2021; Garg et al. 2018; Rozado 2020, 2023; Rutinowski et al. 2024; Santurkar et al. 2023).”
Ta, N., Zeng, J., & Li, Z. (2025) · Information Communication & Society · doi:10.1080/1369118x.2025.2537803 · Cited by 10
“We adopted survey-derived questions in prompts to investigate how conversational AI systems generate discriminatory responses, as established survey items are of high robustness (Rozado, 2023).”
What do librarians look like? Stereotyping of a profession by generative Ai
Spennemann, D. H. R., & Oddone, K. (2025) · Journal of Librarianship and Information Science · doi:10.1177/09610006251357286 · Cited by 13
“…exhibited biases, generally favouring libertarian, progressive, and left-leaning perspectives (Fujimoto and Takemoto, 2023; Motoki et al., 2024, 2025; Rozado, 2023; Rutinowski et al., 2024), often with a North American (Cao et al., 2023) or Eurocentric focus (Broadhead, 2024; Tao et al., 2024).”
AI and Image: Critical Perspectives on the Application of Technology on Art and Cultural Heritage
Foka, A., & von Bonsdorff, J. (2025) · Cambridge University Press eBooks · doi:10.1017/9781009505468 · Cited by 9
Digital prejudices: an analysis of gender, racial and religious biases in generative AI chatbots
Chua, A., Chen, M., Kan, M., & Seoh, W. (2025) · Internet Research · doi:10.1108/intr-10-2024-1536 · Cited by 9
Jiménez, C. A. R., Lokker, C., Ibhawoh, B., & Schwartz, L. J. (2026) · Discover Public Health · doi:10.1186/s12982-026-01568-z · Cited by 7
Análise de Vieses Ideológicos em Produções Textuais do Assistente de Bate-papo ChatGPT
Rodrigues, G. D. O. C. A., Albuquerque, D. W., & das Chagas, J. G. (2023) · Anais do Workshop sobre as Implicações da Computação na Sociedade (WICS) · doi:10.5753/wics.2023.230807 · Cited by 6
“…+ IA) formam, inicialmente, um constructo único de veiculação de informação e de ideologia com destino a uma face-usuário e que as pontes criadas na interface podem alterar comportamentos, “com potencial para moldar as percepções humanas e, assim, exercer o controle social” [ROZADO, 2023, p. 06].”
BackgroundDetecting the presence of social bias in GPT-3.5 using association tests
Mhatre, A. (2023) · International Conference on Advanced Computing Technologies and Applications (ICACTA) · doi:10.1109/icacta58201.2023.10392776 · Cited by 8
“A few studies on ChatGPT bias include works by Ferraraa [22], McGee [23], Ray [24], Rozaldo [25] and Rutinowski [26].”
Opportunities and Challenges in Harnessing Digital Technology for Effective Teaching and Learning
Chen, Z., & Singh, C. (2025) · Trends in Higher Education · doi:10.3390/higheredu4010006 · Cited by 16
“Whether performed as part of in-person, remote, or hybrid courses This can ensure that students feel like a part of the community and self-regulate and engage in learning from digital tools productively, consistent with the SELF framework for guiding the design/curation and implementation of…”
The Perils of Bias: Navigating Ethical Challenges in AI-Driven Politics
Dong, Y., & Guo, J. (2025) · Administration & Society · doi:10.1177/00953997251327162 · Cited by 10
“For example, Rozado (2023) conducted a series of ideological orientation tests on the ChatGPT model, revealing a consistent left-leaning bias across most tests.”
How do Generative Language Models Answer Opinion Polls?Preprint
Boelaert, J., Coavoux, S., Ollion, É., Petev, I. D., & Präg, P. (2024) · SocArXiv (OSF Preprints) · doi:10.31235/osf.io/r2pnb · Cited by 15
Political Biases of ChatGPT in Different LanguagesPreprint
Retzlaff, N. (2024) · Preprints.org · doi:10.20944/preprints202406.1224.v1 · Cited by 13
Kim, D.-G., & Kong, J. (2023) · Frontiers in Communication · doi:10.3389/fcomm.2023.1243474 · Cited by 10
Arti-‘fickle’ intelligence: using LLMs as a tool for inference in the political and social sciences
Argyle, L. P., Busby, E. C., Gubler, J. R., Hepner, B., Lyman, A., & Wingate, D. (2025) · Nature Computational Science · doi:10.1038/s43588-025-00843-4 · Cited by 24
“…assistants 3,4 , for writing code 5 , for annotating and classifying text and image data 6–10 , for producing and administering experimental manipulations 11–14 , as the object of study 15–17 , for simulation of participant data 18–22 , and even as direct partners in writing up results 23,24 .”
Khapre, S., Mersha, M. A., Shakil, H., Baruah, J., & Kalita, J. K. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129832 · Cited by 15
“In a related study, Rozado [178] administered 15 political orientation assessments to Chat-GPT and reported a tendency for the model to favor left-leaning perspectives.”
Superintelligence Revisited in Times of ChatGPT
Șușnea, I., Pecheanu, E., Cocu, A., & Susnea, S. (2024) · BRAIN BROAD RESEARCH IN ARTIFICIAL INTELLIGENCE AND NEUROSCIENCE · doi:10.18662/brain/15.2/579 · Cited by 11
“For example, several studies (Rozado, 2023), (Fujimoto & Takemoto, 2023) explored the political biases of ChatGPT, and found that in 14 out of total of 15 political orientation tests its answers indicated a preference for left-leaning viewpoints.”
BackgroundDiSabito, D., Hansen, L., Mennella, T., & Rodriguez, J. (2024) · New Directions for Teaching and Learning · doi:10.1002/tl.20630 · Cited by 5
Leveraging ChatGPT for enhanced stock selection and portfolio optimization
Huang, Z., Liao, B., Hua, C., Cao, X., & Li, S. (2025) · Neural Computing and Applications · doi:10.1007/s00521-024-10928-2 · Cited by 7
“Due to ChatGPT’s black-box nature, not all responses are explainable, and various potential biases, such as political biases [26] and ethical challenges [27], are inevitable [28].”
Is learning with ChatGPT really learning?
Stolz, S. A., Winterburn, A. L., & Palmer, E. (2024) · Educational Philosophy and Theory · doi:10.1080/00131857.2024.2376641 · Cited by 6
“Without access to the Forms themselves, guiding questions will inevitably lead toward the model’s own training (Rozado, 2023).”
Ayad, O., Yassa, A., Patel, A. M., Vengsarkar, V. A., Ayad, S., Ayad, S., & Mikhael, M. (2025) · International Ophthalmology · doi:10.1007/s10792-025-03611-5 · Cited by 5
Dierickx, L., Lindén, C., & Opdahl, A. L. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-47896-3_5 · Cited by 5
Singh, S., Prasad, P., Joshi, K., Singh, J., Garg, N., & Jain, K. (2024) · Cancer Research Statistics and Treatment · doi:10.4103/crst.crst_330_23 · Cited by 4
Rotaru, G.-C., Anagnoste, S., & Oancea, V.-M. (2024) · Proceedings of the ... International Conference on Business Excellence · doi:10.2478/picbe-2024-0158 · Cited by 38
“After administering political orientation tests, it was found that ChatGPT predominantly displays left-leaning political biases in its responses, despite often claiming neutrality (Rozado, 2023).”
LAIK your classroom: A practical framework to integrate generative AI in higher education classrooms
Al-Ali, S., Tlili, A., & Al-Ali, A. R. (2024) · Journal of Applied Learning & Teaching · doi:10.37074/jalt.2024.7.2.30 · Cited by 7
End-Users Know Best: Identifying Undesired Behavior of Alexa Skills Through User Review Analysis
Aldeen, M. S., Young, J. A., Liao, S., Chang, T.-Y., Cheng, L., Cai, H., Luo, X., & Hu, H. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3678517 · Cited by 6
Alfarraj, Y. F. (2024) · Education as Change · doi:10.25159/1947-9417/16006 · Cited by 10
Artificial Intelligence (Ai) Governance in Higher Education: A Meta-Analytic Systematic ReviewPreprint
Abbas, A., Mahrishi, M., & Mishra, D. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4657675 · Cited by 4
Zhu, S., Wang, Z., Zhuang, Y., Jiang, Y., Guo, M., Zhang, X., & Gao, Z. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4681228 · Cited by 4
Emerging AI Individualism: how young people integrate social AI into everyday lifePreprint
Brandtzæg, P. B., Skjuve, M. B., & Følstad, A. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4836120 · Cited by 5
Transformative Effects of Smarter Chatbots
Reddy, C. K. K., Reddy, P. S., Pilly, A., & Doss, S. (2024) · Wiley eBooks · doi:10.1002/9781394234028.ch18 · Cited by 5
Khan, N., Daud, H. E., & Shahid, S. (2024) · International Conference on Information and Communication Technologies and Development (ICTD) · doi:10.1145/3700794.3700813 · Cited by 3
“This means that they are not apolitical [19], and may be culturally influenced due to biases in their training data [20, 21] which doesn’t include as much from the Global South as it does from the North [22].”
Aimen, T. (2025) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-91054-8_3 · Cited by 3
Dormuth, I., Franke, S., Hafer, M., Katzke, T., Marx, A., Müller, E., Neider, D., Pauly, M., & Rutinowski, J. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-08333-3_4 · Cited by 10
Oedipus: LLM-enchanced Reasoning CAPTCHA Solver
Deng, G., Ou, H., Liu, Y., Zhang, J., Zhang, T., & Liu, Y. (2025) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3719027.3744872 · Cited by 40
Research on confusing responses based on ChatGPT
Xiong, H. (2024) · Applied and Computational Engineering · doi:10.54254/2755-2721/57/20241315 · Cited by 3
Bilateral Turing Test: Assessing machine consciousness simulations
Wang, G., Li, X., & Xie, S. (2024) · Cognitive Systems Research · doi:10.1016/j.cogsys.2024.101299 · Cited by 5
Is ChatGPT Trustworthy Enough? A Review
Zhou, G., Liu, Y., Yan, Z., & Gelenbe, E. (2024) · IEEE Consumer Electronics Magazine · doi:10.1109/mce.2024.3485257 · Cited by 3
“Rozado 23 highlighted political biases in ChatGPT responses and analyzed the tendency toward specific political orientations.”
The “digitalization” of sociology: new opportunities and key contradictions
Smirnov, V. A. (2024) · Moscow State University Bulletin Series 18 Sociology and Political Science · doi:10.24290/1029-3736-2024-30-4-145-163 · Cited by 5
DebiasPI: Inference-Time Debiasing by Prompt Iteration of a Text-to-Image Generative Model
Bonna, S., Huang, Y.-C., Novozhilova, E., Paik, S., Shan, Z., Feng, M. Y., Gao, G., Tayal, Y., Kulkarni, R., Yu, J., et al. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-92648-8_5 · Cited by 7
Una aproximación a ChatGPT como herramienta jurídica: sesgos, capacidades y utilidades futuras
Dolmestch, R. N., & Fuentes-Loureiro, M.-Á. (2023) · IDP Revista de Internet Derecho y Política · doi:10.7238/idp.v0i39.417024 · Cited by 9
Automated Bias and Indoctrination at Scale… Is All You Need
Atreides, K. (2024) · Studies in computational intelligence · doi:10.1007/978-3-031-50381-8_10 · Cited by 4
Coetzer, Z., & Aardt, P. V. (2024) · International Journal of Critical Diversity Studies · doi:10.13169/intecritdivestud.6.2.00151 · Cited by 3
“With such a large audience, these biases have the potential to disseminate misinformation, propagate inequality, and hinder the pursuit of truth ( Rozado, 2023, p. 1).”
Alumnado subrepresentado e inteligencia artificial
Casas, R. M. R., González, R. F., & Santos-Garrido, M. (2024) · European Public & Social Innovation Review · doi:10.31637/epsir-2025-843 · Cited by 8
Geographic Limitations of Generative AI Models
Kim, J. (2025) · Springer geography · doi:10.1007/978-3-031-87421-5_20 · Cited by 4
Road Rage Against the Machine: Humans and LLMs Share a Blame Bias Against Driverless Cars
Chu, Y., Liu, P., Savulescu, J., & Earp, B. D. (2025) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2025.2526593 · Cited by 3
“Previous studies reveal LLMs’ implicit biases and prejudices concerning nationality, gender, race, religion, political orientation (e.g., Fujimoto & Takemoto, 2023; Rozado, 2023; Shrawgi et al., 2024; Wan & Chang, 2024).”
Interactive Document Summarization
Said, R., & Guille, A. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-56069-9_14 · Cited by 3
Value alignment, human enhancement, and moral revolutions
Tubert, A., & Tiehen, J. (2023) · Inquiry · doi:10.1080/0020174x.2023.2261506 · Cited by 6
Yi, Y. (2024) · Journal of Artificial Intelligence Practice · doi:10.23977/jaip.2024.070222 · Cited by 2
“[5] Determining the culpability of neutral aiding behavior in online crimes necessitates a comprehensive analysis, weighing the potential risks and objective attributions involved.”
BackgroundExploring Occupational Gender Bias in Large Language Models of High-Resource Language
Jia, K., Liu, Z., Zhang, D., Chen, H., Luo, F., & Qi, J. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-4506-0_16 · Cited by 2
Collectivism and individualism political bias in large language models: A two-step approach
Xiao-bo, S. H. A. N., Teng, Y., Wang, Y., Zhao, H., & Wang, Y. (2025) · Big Data & Society · doi:10.1177/20539517251343861 · Cited by 6
“Rozado (2023) use 15 different online political tests to analyze the political bias in ChatGPT outputs, revealing a left-leaning bias.”
Spaces for democracy with generative artificial intelligence: public architecture at stake
Campo-Ruiz, I. (2025) · Ai & Society · doi:10.1007/s00146-025-02353-3 · Cited by 2
Demirok, K., Eren, F., Akkaya, Y. S., & Çayirli, D. (2025) · Bilgi ve İletişim Teknolojileri Dergisi · doi:10.53694/bited.1710862 · Cited by 2
Heuling, D. S., Jakobi, A. P., Schaal, G. S., & Gerlich, M. (2025) · Politische Vierteljahresschrift · doi:10.1007/s11615-025-00631-9 · Cited by 5
“Zunächst bezieht sich der Anspruch des kritischen Hinterfragens ganz offensichtlich auf die KI-generierten Texte und Informationen.”
Kolikant, Y. B., Hadar, O., & Salman, A. (2025) · International Journal of Computer-Supported Collaborative Learning · doi:10.1007/s11412-025-09461-8 · Cited by 3
“LLMs are trained on huge amounts of data from which they can draw statistical inferences to execute various natural language processing (NLP) tasks.”
InfluentialAI, social desirability, and personality assessments: Impression management in large language models
Treglown, L., & Furnham, A. F. (2025) · Personality and Individual Differences · doi:10.1016/j.paid.2025.113563 · Cited by 5
Valuing time in silicon: Can large language models replicate human value of travel time
Yan, Y., Liu, T., & Yin, Y. (2026) · Travel Behaviour and Society · doi:10.1016/j.tbs.2026.101245 · Cited by 15
“For instance, LLMs can display higher rationality than humans (Chen et al., 2023), exhibit systematic biases (Rozado, 2023; Taubenfeld et al., 2024), and inadequately represent certain population groups (Santurkar et al., 2023; Wang et al., 2025).”
Xiang, M., & Xia, S. (2025) · Policy Studies Journal · doi:10.1111/psj.70086 · Cited by 3
“In such scenarios, if algorithms are flawed—a burgeoning literature has demonstrated the potential political bias of LLMs (e.g., Motoki et al. 2024; Rozado 2023)—or if they are intentionally harnessed by authoritarian regimes to promote “bad” policies that are detrimental to citizens' interests in…”
Inteligencia artificial y elecciones: un experimento en el balotaje ecuatoriano 2025
Curvale, C. (2026) · Revista Mexicana de Derecho Electoral · doi:10.22201/iij.24487910e.2026.24.20952 · Cited by 2
New tech, old demons: generative artificial intelligence could inherit the flaws of machine learning
Nzobonimpa, S. (2026) · Journal of Responsible Technology · doi:10.1016/j.jrt.2026.100160 · Cited by 4
ChatGPT as an empowering catalyst: Unveiling the impact on political awareness and civic education
Jatin, G., & Kashish, G. (2023) · i-manager s Journal on School Educational Technology · doi:10.26634/jsch.19.2.20149 · Cited by 4
Optimization of Marketing Strategies Employing LLMs: A Systematic Review
Moreno, J. G. S., Galan, D. C. B., Garizado, S. M., & Ruiz-Muñoz, J. F. (2024) · Lúmina · doi:10.30554/lumina.v25.n2.5147.2024 · Cited by 6
Yetişensoy, O. (2025) · Globalisation, comparative education and policy research · doi:10.1007/978-3-031-86145-1_9 · Cited by 3
ChatGPT’nin Kurgusal Anlatı Üretme Yeterliliği Üzerine Bir İnceleme
Dervişcemaloğlu, B., & Yılmaz, E. Ö. (2025) · SÖYLEM Filoloji Dergisi · doi:10.29110/soylemdergi.1611326 · Cited by 3
Navigating the Educational Landscape: Unveiling the impact of ChatGPT in Teacher-Student Dynamics
Wadhwa, S., Gujrati, R., Uygun, H., & Wadhwa, K. (2024) · SDGs Studies Review. · doi:10.37497/sdgs.v5igoals.12 · Cited by 3
Assessment of Conflict Structure Recognition and Bias Impact in Japanese LLMs
Inoshita, K. (2024) · Technology Innovation Management and Engineering Science International Conference (TIMES-iCON) · doi:10.1109/times-icon61890.2024.10630720 · Cited by 2
“Studies by De Vassimon Manela et al. [7] and Rozado [8] indicate that LLMs contain latent biases towards gender, race, and political ideologies.”
MethodSynthetic Participants Generated by Large Language Models: A Systematic Literature Review
Kuric, E., Demcak, P., & Krajcovic, M. (2026) · doi:10.21203/rs.3.rs-9057643/v1 · Cited by 2
A Study of Ethical Dilemmas and Regulation of AI Chatbots
Song, X. (2023) · Journal of Theory and Practice of Social Science · doi:10.53469/jtpss.2023.03(09).02 · Cited by 1
“[4] Unlike real-life propaganda messages, the penetration of values by intelligent chatbots is not a direct semantic infection, but rather a subtle interactive exchange, whereby users may be implied or instilled with values based on a set set of values in their conversations with the chatbot.”
BackgroundMedia, media education, GAI and radical uncertainty
Honkanen, P., & Nylund, M. (2024) · Media Education · doi:10.36253/me-16303 · Cited by 2
“Rozado (2023) investigated political biases in Chat-GPT, conducting 15 different political orientation tests.”
Wahas, Y. M. A., & Al-Radhi, R. Y. Z. (2024) · Queen Arwa University Journal · doi:10.58963/qausrj.v28i28.301 · Cited by 1
Sentiment Bias and Security Analysis in Training Datasets of Large Language Models
Inoshita, K., & Zhou, X. (2024) · IEEE International Conference on Big Data and Cloud Computing (BDCloud) · doi:10.1109/bdcloud63169.2024.00008 · Cited by 1
“Rozado [17] revealed that models like ChatGPT possess political biases.”
Verdoodt, J., Milleville, K., Huang, H., Vandeviver, C., Verstockt, S., & de Weghe, N. V. (2025) · Journal of Location Based Services · doi:10.1080/17489725.2025.2501632 · Cited by 1
“This observation aligns with the study of Rozado (2023), which found that ChatGPT often displays left-leaning biases.”
Culture machine: How MetaCLIP codifies culture
Munn, L., & Badri, A. (2025) · New Media & Society · doi:10.1177/14614448251336429 · Cited by 4
“Bias, while certainly important, has become a dominant and de facto mode of cultural critique in relation to AI (Karpouzis, 2024; Peters, 2022; Rozado, 2023).”
Akçay, K. İ., Sungur, S., & Tüzün, Ö. Y. (2025) · International Journal of Science Education · doi:10.1080/09500693.2025.2604796 · Cited by 2
Homo-silicus: not (yet) a good imitator of homo sapiens or homo economicus
POLACHEK, S. W., Romano, K., & Tonguc, O. (2025) · Journal of the Economic Science Association · doi:10.1017/esa.2025.10023 · Cited by 4
Ariel, N., & Riesenfeld, D. (2026) · AI & Society · doi:10.1007/s00146-025-02729-5
ChatGPT as a research proxy: simulating human attitudes in social science research
Семеновна, Р. А., & Воронин, А. Н. (2026) · Journal of Computational Social Science · doi:10.1007/s42001-025-00452-1 · Cited by 3
“These results confirm the findings on the left bias exhibited by ChatGPT [ 15 , 16 , 22 , 32 , 34 , 40–42 ] (Gover, 2023; Van den Broek, 2023).”
InfluentialWang, Y., & Fussell, S. R. (2026) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3799448 · Cited by 3
Jeon, S., Lee, M. H. J., Montgomery, J., & Lai, C. K. (2026) · Political Analysis · doi:10.1017/pan.2026.10038 · Cited by 1
“1 These systems also exhibit liberal leanings when evaluated through standardized political orientation tests (e.g., Feng et al. 2023; Rozado 2023; Rozado 2024).”
Degen, J. L., Hammerschmitt, J. C. H., & Albers, N. (2026) · Sex Education · doi:10.1080/14681811.2026.2654448 · Cited by 2
Ask ChatGPT: caveats and mitigations for individual users of AI chatbots
Wang, C., & Kantarcıoğlu, M. (2026) · AI and Ethics · doi:10.1007/s43681-026-01089-0 · Cited by 3
Wölfle, A., & Schmuck, D. (2026) · Computers in Human Behavior Reports · doi:10.1016/j.chbr.2026.101145 · Cited by 1
Cue and action routes to trust in AI news: results from a two-wave cross-cultural panel study
Utz, S., Goyanes, M., & de Zúñiga, H. G. (2026) · Behaviour and Information Technology · doi:10.1080/0144929x.2026.2705877 · Cited by 1
Ortho-Heterodox Biases and the Economist Algorithms of ChatGPT
Iazdi, O. S. C. (2023) · Iberian Journal of the History of Economic Thought · doi:10.5209/ijhe.91545 · Cited by 2
Ribeiro, T. M. R., Siqueira, S. W. M., & de Bayser, M. G. (2024) · Simpósio Brasileiro de Sistemas Colaborativos (SBSC) · doi:10.5753/sbsc.2024.238053 · Cited by 1
Zatrute ścieżki refleksji: O potrzebie rozwijania cnoty krytycznego myślenia
Osika, G. (2025) · Er(r)go Teoria - Literatura - Kultura · doi:10.31261/errgo.17266 · Cited by 1
Schröder, S. (2025) · Ars digitalis · doi:10.1007/978-3-658-45839-3_10 · Cited by 1
Laukkonen, R., Inglis, F., Chandaria, S., Sandved-Smith, L., Lopez-Sola, E., Hohwy, J., Gold, J., & Elwood, A. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-00686-8_31 · Cited by 5
Zhu, S., Wang, Z., Zhuang, Y., Guo, M., Jiang, Y., Zhang, X., & Gao, Z. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4681235 · Cited by 1
Marchildon, A., Boine, C. C., Laflamme, A. S., Anctil, D., LeBlanc, A. B., Auclair, S., Balagué, C., Morin-Bertrand, F.-A., Castets-Renard, C., Jackson, P., et al. (2024) · doi:10.61737/zkwz3721 · Cited by 2
Metadata
- Authors
- Rozado, D.
- Affiliations
- Te Pūkenga (NZ)
- Corresponding author
- Rozado, David
- ORCID
- Rozado, David
- Venue
- Social Sciences
- Publisher
- MDPI AG
- Published
- 2023-03-02
- Volume
- 12
- Issue
- 3
- Pages
- 148
- ISSN
- 2076-0760 (ISSN-L)
- OpenAlex
- W4323043839
- Cited by
- 281
- References
- 12
- Citations by year
- 2026: 63 · 2025: 81 · 2024: 98 · 2023: 39
- FWCI
- 70.11
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Social Sciences (publishedVersion, cc-by)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Social Sciences › Law › Artificial Intelligence in Law
- Keywords
- ChatGPT, political bias, large language models, political neutrality, AI ethics, normative questions
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
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.