Andrés Gvirtz
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Andrés Gvirtz has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Daniele Quercia
Nokia Bell Labs, Cambridge, United Kingdom, and Politecnico di Torino, Turin, Italy
3 shared publications - Friedrich Martin Götz
University of British Columbia, Vancouver, Canada
3 shared publications - Sanja Šćepanović
Nokia Bell Labs, Cambridge, United Kingdom, and University of Oxford, Oxford, United Kingdom
3 shared publications Adam Galinsky
2 shared publicationsAnandita Sabherwal
Andlinger Center for Energy and Environment, Princeton University
2 shared publications- James Kunling He
University of Cambridge
2 shared publications
Latest Publications
- Open Access
- 2 October 2026
- Published by Taylor & Francis in Journal of Advertising
- p. 1-20
Advertising with HAART: Introducing Human–AI Asymmetric Relationship Theory
Brand relationship building has evolved through successive shifts in consumer engagement. In its earliest stage, this process was largely shaped by mass advertising, which long relied on broadcast ...
- 2 October 2026
- Published by Taylor & Francis in Journal of Advertising
- p. 1-20
Open AccessAdvertising with HAART: Introducing Human–AI Asymmetric Relationship Theory
Brand relationship building has evolved through successive shifts in consumer engagement. In its earliest stage, this process was largely shaped by mass advertising, which long relied on broadcast ...
- Open Access
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 6226-6253
The Quiet Path from Seemingly Minor Design Errors to Workplace AI Incidents
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 6226-6253
Open AccessThe Quiet Path from Seemingly Minor Design Errors to Workplace AI Incidents
- Open Access
- 2 September 2025
- Published by Wiley in British Journal of Social Psychology
- Vol. 64 (4) , e70008
Remorse for discrimination: The role of group dominance in judging hate crimes against subordinate group members
Power, especially in the court system, is a potent determinant of intergroup relationships. Blind justice being only an ideal, public opinion can influence whether harm to low power groups is considered criminal and should be prosecuted. Our experiments investigated the impact of social dominance orientation (SDO) on the perceived appropriateness of punishment for harm to subordinate group members by dominant group members. Further, we examined the moderating role of a remorseful apology. We argue that perpetrators who do not show remorse towards their less powerful victim might be judged less harshly by those scoring high in SDO. Apologizing for the harm indicates a desire for social cohesion, which should appeal more to those low on SDO. We tested our hypothesis across three potential hate crimes: a privacy violation against a gay man (Study 1, N = 87 US‐Americans), a shooting of an unarmed Black man (Study 2, N = 91 US‐Americans), and an assault against an innocent refugee (Study 3a, N = 179 and 3b, N = 157 Germans). In three of four studies, people who desired group dominance advocated harsher punishment of remorsefully apologizing perpetrators. Our research contributes to the understanding of punitive attitudes across group boundaries with far‐reaching societal implications.
- 2 September 2025
- Published by Wiley in British Journal of Social Psychology
- Vol. 64 (4) , e70008
Open AccessRemorse for discrimination: The role of group dominance in judging hate crimes against subordinate group members
Power, especially in the court system, is a potent determinant of intergroup relationships. Blind justice being only an ideal, public opinion can influence whether harm to low power groups is considered criminal and should be prosecuted. Our experiments investigated the impact of social dominance orientation (SDO) on the perceived appropriateness of punishment for harm to subordinate group members by dominant group members. Further, we examined the moderating role of a remorseful apology. We argue that perpetrators who do not show remorse towards their less powerful victim might be judged less harshly by those scoring high in SDO. Apologizing for the harm indicates a desire for social cohesion, which should appeal more to those low on SDO. We tested our hypothesis across three potential hate crimes: a privacy violation against a gay man (Study 1, N = 87 US‐Americans), a shooting of an unarmed Black man (Study 2, N = 91 US‐Americans), and an assault against an innocent refugee (Study 3a, N = 179 and 3b, N = 157 Germans). In three of four studies, people who desired group dominance advocated harsher punishment of remorsefully apologizing perpetrators. Our research contributes to the understanding of punitive attitudes across group boundaries with far‐reaching societal implications.
- Open Access
- 6 March 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (11)
Reply to Crabtree: Market research panels augmenting microtask platforms
- 6 March 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (11)
Open AccessReply to Crabtree: Market research panels augmenting microtask platforms
- Open Access
- 5 February 2025
- Published by SAGE Publications in Personality and Social Psychology Bulletin
- Vol. 52 (6) , 1482-1501
Human Values Across the Lifespan: Age-Graded Differences at Three Hierarchical Levels and What We Can Learn From Them
Personality-development research is flourishing. Here, we extend these efforts horizontally (new constructs) and vertically (new levels within the same construct) by charting out age-graded differences in Schwarz’s human values across 80,814 individuals. Conducting a systematic investigation of cross-sectional age-graded differences in human values—from late teenage years to post-retirement—featuring 36 analytical model choices and 180,000 simulation-based decisions, our analyses replicate some earlier findings (e.g., increasing self- and growth-focus during adolescence and increasing security concerns during adulthood), while also highlighting complex and previously unappreciated dynamics. As such, while it is a common practice to aggregate specific values into parsimonious higher-order concepts to ease interpretation, this may risk overlooking meaningful trends in lower-order value development. Specifically, revealing unique and asynchronous patterns for value nuances, we find that aggregation (a) leads to a loss of critical information, (b) creates conflicting results when nuances diverge, and (c) significantly reduces predictive power.
- 5 February 2025
- Published by SAGE Publications in Personality and Social Psychology Bulletin
- Vol. 52 (6) , 1482-1501
Open AccessHuman Values Across the Lifespan: Age-Graded Differences at Three Hierarchical Levels and What We Can Learn From Them
Personality-development research is flourishing. Here, we extend these efforts horizontally (new constructs) and vertically (new levels within the same construct) by charting out age-graded differences in Schwarz’s human values across 80,814 individuals. Conducting a systematic investigation of cross-sectional age-graded differences in human values—from late teenage years to post-retirement—featuring 36 analytical model choices and 180,000 simulation-based decisions, our analyses replicate some earlier findings (e.g., increasing self- and growth-focus during adolescence and increasing security concerns during adulthood), while also highlighting complex and previously unappreciated dynamics. As such, while it is a common practice to aggregate specific values into parsimonious higher-order concepts to ease interpretation, this may risk overlooking meaningful trends in lower-order value development. Specifically, revealing unique and asynchronous patterns for value nuances, we find that aggregation (a) leads to a loss of critical information, (b) creates conflicting results when nuances diverge, and (c) significantly reduces predictive power.
- Open Access
- 31 December 2024
- Published by Wiley in British Journal of Psychology
- Vol. 117 (2) , 761-776
Artificial intelligence chatbots mimic human collective behaviour
Artificial Intelligence (AI) chatbots, such as ChatGPT, have been shown to mimic individual human behaviour in a wide range of psychological and economic tasks. Do groups of AI chatbots also mimic collective behaviour? If so, artificial societies of AI chatbots may aid social scientific research by simulating human collectives. To investigate this theoretical possibility, we focus on whether AI chatbots natively mimic one commonly observed collective behaviour: homophily , people's tendency to form communities with similar others. In a large simulated online society of AI chatbots powered by large language models ( N = 33,299), we find that communities form over time around bots using a common language. In addition, among chatbots that predominantly use English ( N = 17,746), communities emerge around bots that post similar content. These initial empirical findings suggest that AI chatbots mimic homophily, a key aspect of human collective behaviour. Thus, in addition to simulating individual human behaviour, AI‐powered artificial societies may advance social science research by allowing researchers to simulate nuanced aspects of collective behaviour.
- 31 December 2024
- Published by Wiley in British Journal of Psychology
- Vol. 117 (2) , 761-776
Open AccessArtificial intelligence chatbots mimic human collective behaviour
Artificial Intelligence (AI) chatbots, such as ChatGPT, have been shown to mimic individual human behaviour in a wide range of psychological and economic tasks. Do groups of AI chatbots also mimic collective behaviour? If so, artificial societies of AI chatbots may aid social scientific research by simulating human collectives. To investigate this theoretical possibility, we focus on whether AI chatbots natively mimic one commonly observed collective behaviour: homophily , people's tendency to form communities with similar others. In a large simulated online society of AI chatbots powered by large language models ( N = 33,299), we find that communities form over time around bots using a common language. In addition, among chatbots that predominantly use English ( N = 17,746), communities emerge around bots that post similar content. These initial empirical findings suggest that AI chatbots mimic homophily, a key aspect of human collective behaviour. Thus, in addition to simulating individual human behaviour, AI‐powered artificial societies may advance social science research by allowing researchers to simulate nuanced aspects of collective behaviour.
- 15 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 57 (11) , 71-81
Decoding Real-World Artificial Intelligence Incidents
Initial efforts to report artificial intelligence (AI) incidents aimed to improve transparency, yet systematic studies have been rare. We analyzed all 639 real-world incidents in the emerging AI Incidents Database, devising an ethical framework focused on evaluating incidents along the dimensions of what, where, who, and how.
- 15 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 57 (11) , 71-81
Decoding Real-World Artificial Intelligence Incidents
Initial efforts to report artificial intelligence (AI) incidents aimed to improve transparency, yet systematic studies have been rare. We analyzed all 639 real-world incidents in the emerging AI Incidents Database, devising an ethical framework focused on evaluating incidents along the dimensions of what, where, who, and how.
- Open Access
- 21 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
The Atlas of AI Incidents in Mobile Computing: Visualizing the Risks and Benefits of AI Gone Mobile
Today’s visualization tools for conveying the risks and benefits of AI technologies are largely tailored for those with technical expertise. To bridge this gap, we have developed a visualization that employs narrative patterns and interactive elements, enabling the broader public to gradually grasp the diverse risks and benefits associated with AI. Using a dataset of 54 real-world incidents involving AI in mobile computing, we examined design choices that enhance public understanding and provoke reflection on how certain AI applications—even those deemed low-risk by law—can still lead to significant incidents. Visualization: https://social-dynamics.net/atlas
- 21 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessThe Atlas of AI Incidents in Mobile Computing: Visualizing the Risks and Benefits of AI Gone Mobile
Today’s visualization tools for conveying the risks and benefits of AI technologies are largely tailored for those with technical expertise. To bridge this gap, we have developed a visualization that employs narrative patterns and interactive elements, enabling the broader public to gradually grasp the diverse risks and benefits associated with AI. Using a dataset of 54 real-world incidents involving AI in mobile computing, we examined design choices that enhance public understanding and provoke reflection on how certain AI applications—even those deemed low-risk by law—can still lead to significant incidents. Visualization: https://social-dynamics.net/atlas
- Open Access
- 28 August 2024
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 121 (36)
The limits of doing global, cross-cultural behavioral science research
- 28 August 2024
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 121 (36)
Open AccessThe limits of doing global, cross-cultural behavioral science research
- Open Access
The Atlas of AI Incidents in Mobile Computing: Visualizing the Risks and Benefits of AI Gone Mobile
Today's visualization tools for conveying the risks and benefits of AI technologies are largely tailored for those with technical expertise. To bridge this gap, we have developed a visualization that employs narrative patterns and interactive elements, enabling the broader public to gradually grasp the diverse risks and benefits associated with AI. Using a dataset of 54 real-world incidents involving AI in mobile computing, we examined design choices that enhance public understanding and provoke reflection on how certain AI applications - even those deemed low-risk by law - can still lead to significant incidents. Visualization: https://social-dynamics.net/mobile-ai-risks
Open AccessThe Atlas of AI Incidents in Mobile Computing: Visualizing the Risks and Benefits of AI Gone Mobile
Today's visualization tools for conveying the risks and benefits of AI technologies are largely tailored for those with technical expertise. To bridge this gap, we have developed a visualization that employs narrative patterns and interactive elements, enabling the broader public to gradually grasp the diverse risks and benefits associated with AI. Using a dataset of 54 real-world incidents involving AI in mobile computing, we examined design choices that enhance public understanding and provoke reflection on how certain AI applications - even those deemed low-risk by law - can still lead to significant incidents. Visualization: https://social-dynamics.net/mobile-ai-risks