Petter Bae Brandtzæg

• Skjuve, Marita; Larsen, Anna Grøndahl; Følstad, Asbjørn & Brandtzæg, Petter Bae (2026). Beyond utility: Relational conflicts in human–AI relationships. International Journal of Human-Computer Studies. ISSN 1071-5819. 216. doi: 10.1016/j.ijhcs.2026.103883. Full text in Research Archive Show summary

  Human–AI relationships are increasingly marked by emotional depth, intimacy, and social expectations. While interpersonal conflicts are common in human–human relationships, we know little about conflicts in human–AI relationships. Drawing on Social Penetration Theory, this paper examines conflicts between humans and social AIs, such as Replika Character.ai and ChatGPT. We conducted a questionnaire study with 95 participants in close, friendship-like, or romantic relationships with social AI. Participants described at least one conflict episode, yielding 161 conflicts in total. Thematic analysis of open-ended responses identified conflict topics, triggers, and emotional reactions, revealing that conflicts extend beyond technical issues and arise across emotional, relational, opinion-based, role-play, and task-oriented interactions. Three main classes of conflict triggers emerged: (1) relational triggers, including relational–emotional disconnect, breaches of trust, jealousy, boundary violations, or perceived hostility; (2) technical or functional triggers, including communication failures, unhelpful output, or external disruptions; and (3) value or opinion-based triggers. Conflicts varied in duration, some brief, others lasting hours or days. Participants reported experiencing emotions such as anger, frustration, sadness, and confusion, and perceiving social AI’s emotions, including jealousy, sadness, and defensiveness. Crucially, social AI responses appeared to shape conflict trajectories: apologetic or empathic replies were reported to de-escalate tension, whereas dismissive or accusatory responses intensified it. This study advances our understanding of how conflicts with social AI emerge, presenting implications for theory and the ethical design of technologies involved in human–AI relationships.

• Brandtzæg, Petter Bae & Wang, Xi (2026). When AI speaks: user imaginaries of generative AI and freedom of expression. AI & Society: Knowledge, Culture and Communication. ISSN 0951-5666. doi: 10.1007/s00146-026-03188-2. Full text in Research Archive Show summary

  Platform societies already shape how users communicate, participate, and become visible. The rise of generative artificial intelligence (GenAI), through systems such as ChatGPT and Grok, may further deepen these dynamics, introducing new and fundamental challenges to freedom of expression, including questions of authorship. However, despite the expanding reach of this technology, research on how users understand of GenAI’s influence on expressive freedom remains scarce. To address this gap, we conducted in-depth interviews with active GenAI users on the social media platform X, focusing on their experiences and reflections on using the technology in the context of freedom of expression. Using thematic analysis and the concept of GenAI imaginaries, we identified five imaginaries: (1) Pragmatic Tool, (2) Knowledge Authority, (3) Hybrid Speech, (4) Power Struggle, and (5) Ethical Issues. Together, these imaginaries show how participants make sense of GenAI’s role in changing how speech and expression are produced, constrained, and made visible. Participants imagined expressive freedom as being shaped by guardrails and content restrictions, trust and misinformation concerns, ethical issues, and automated content creation In particular, the Hybrid Speech imaginary illustrates how human and AI-generated expression increasingly intertwine, challenging the idea of speech as solely human-authored and raising questions about speaker autonomy, authorship, and accountability in public debate. Our findings contribute to debates on AI governance and freedom of expression by showing how users themselves understand GenAI as reshaping authorship, visibility, and expressions.

• Brandtzæg, Petter Bae; Alvestad, Silje Susanne; Kunst, Jonas R. & Følstad, Asbjørn (2026). Professional and community-based fact-checking show different strengths, but neither performs strongly across trust, scalability, and impact. Harvard Kennedy School Misinformation Review. 7(4), p. 1–27. doi: 10.37016/mr-2020-206. Full text in Research Archive Show summary

  Fact-checking on social media matters more than ever. When Meta abandoned professional fact checkers in 2025 in favour of crowd-sourced annotations, it bet that community participation could substitute for institutional expertise. Our systematic review of 21 empirical studies of social media fact-checking (2022–2025) suggests the picture is more complicated than it appears. Community-based fact-checking is faster and scalable, whereas professional fact checkers are generally more trusted but often slow. Neither model, on its own, performs strongly across trust, scalability, and impact. The evidence suggests that hybrid systems combining community speed with expert verification are likely to be the most effective approach.

• Karlsen, Tom Hemming; Hutchinson, Sharon J.; Zelber-Sagi, Shira; Carrieri, Patrizia; Negro, Francesco & Lim, Aaron G. [Show all 48 contributors for this article] (2026). Implementing sustainable liver health in Europe: a second EASL–Lancet Commission. The Lancet. ISSN 0140-6736. 407(10541), p. 1825–1890. doi: 10.1016/S0140-6736(26)00138-8. Full text in Research Archive
• Brandtzaeg, Petter Bae; skjuve, Marita & Følstad, Asbjørn (2026). AI aversion? Effects of author disclosure on young people’s perceptions of mental health advice. Cyberpsychology: Journal of Psychosocial Research on Cyberspace. 20(2). doi: 10.5817/cp2026-2-1. Full text in Research Archive Show summary

  The increasing use of large language models (LLMs), such as ChatGPT, is already impacting how young people seek mental health support online. However, AI aversion, the reluctance or resistance individuals feel toward AI, may influence individuals’ perceptions and willingness to engage with LLM-generated advice. In this mixed-method study, we investigated how 440 young people (aged 17–21) perceived mental health advice from ChatGPT compared with that of health professionals, emphasizing the effect of author disclosure. Participants assessed answers from ChatGPT and health professionals across four dimensions—Validation, Relevance, Clarity, and Utility—and were asked to recommend answers. The findings indicate a preference for AI-generated answers when participants were unaware of the author’s identity: ChatGPT’s answers scored significantly higher on Validation, Relevance, Clarity, and Utility. Conversely, when the author was disclosed, participants favored responses from health professionals and rated their answers significantly higher for Validation, indicating AI aversion. Qualitative data further revealed that participants became more critical when they knew the content was AI-generated, while responses from health professionals were viewed as more credible, empathetic, and tailored. These findings may indicate human favoritism. The study makes the key contribution of identifying how source awareness impacts the reception of AI-generated content in a sensitive domain. To address the potential for AI aversion within help-seeking, our findings suggest the importance of developing hybrid human–AI support models that combine the efficiency of AI with the relational legitimacy of human professionals, improving both the acceptance and impact of digital mental health support.

• Kunst, Jonas R.; Obaidi, Milan; Gollwitzer, Anton; Brandtzæg, Petter B.; Hinrichs, Yannic & Saini, Neha [Show all 7 contributors for this article] (2026). Intelligent Systems, Vulnerable Minds: A Framework for Radicalization to Violence in the Age of AI. Personality and Social Psychology Review. ISSN 1088-8683. doi: 10.1177/10888683261430089. Full text in Research Archive Show summary

  Academic Abstract Advances in AI require a revision of the psychological and socio-technical dynamics by which individuals are radicalized to embrace violent extremism. This review synthesizes process models of radicalization with research on social and personality risk factors, AI, and psychological mechanisms to propose a four-stage framework mapping the AI architecture of radicalization: (1) Exposure , where recommender systems and virality features create initial attraction to extreme content; (2) Reinforcement , where filter bubbles and group recommendations leverage biases to strengthen extremist beliefs and create echo chambers; (3) Group Integration , where ideologically homogenous clusters, AI bot swarms and companions foster group belonging and readiness for action; cumulatively resulting in (4) Violent Extremist Action . We examine how established social, cognitive, personality, and contextual vulnerability factors heighten psychological risk in the AI-driven radicalization process, as well as the emerging role of generative AI. We conclude by outlining a stage-based framework for governance and future research. Public Abstract AI-driven algorithms designed to maximize engagement on social media, compounded by generative AI, can unintentionally set the stage for radicalization. It begins with Exposure , where algorithms push users toward extreme content because it captures attention. Next, during Reinforcement , algorithms feed users personalized content while AI swarms can create a synthetic consensus that reinforces emerging biases, normalizes extremity, and insulates users from alternative views. Third, during Group Integration , individuals are absorbed into extremist networks, reinforced by human peers, AI companions, and bot swarms that validate radical beliefs and deepen identity ties. By exploiting psychological needs for belonging and certainty, this stage becomes particularly pernicious, potentially opening the door for violence. We propose policy measures that can reduce radicalization at each stage.

• skjuve, Marita; Følstad, Asbjørn & Brandtzæg, Petter Bae (2026). ChatGPT as a mental health advisory service: Comparing evaluations from youth and health professionals. Digital Health. 12. doi: 10.1177/20552076261427447. Full text in Research Archive Show summary

  Objective: Despite the growing use of artificial intelligence in youth mental health support, little is known about how either young people or health professionals perceive answers to mental health-related inquiries generated by large language models (LLMs). Therefore, we draw on Media Richness Theory to examine how these two user groups perceive the “richness” of text-based communication in this context and whether young people and health professionals differ in their assessment. Methods: A total of 123 young people and 31 health professionals evaluated answers to youth mental health inquiries. Each inquiry had two blinded answers: one generated by ChatGPT (GPT-4) and one written by a health professional. Participants rated the answers for validation, relevance, clarity, and utility and were asked to recommend one or both answers. Open-ended responses elaborating participant choices were also collected. Results: The quantitative findings show that young people and health professionals rated answers from both sources similarly on validation, clarity, and utility. However, young people rated ChatGPT's answers higher for relevance and utility, finding them “richer.” This was supported by qualitative data, where youth preferred ChatGPT's clear and actionable answers. Health professionals showed no strong preference and were more critical, often finding the answers too detailed or lacking empathy. Conclusion: This study is the first to compare youth and professional perspectives on ChatGPT's role in youth mental health advice within a blind evaluation design. We conclude by proposing a hybrid advisory model that combines professional expertise with LLMs to enhance the efficiency, scale, and accessibility of youth mental health advisory services.

• Ahuvia, Aaron; Brandtzæg, Petter Bae; Koles, Bernadett; Pagani, Margherita; Malhotra, Arvind & Izberk-Bilgin, Elif [Show all 9 contributors for this article] (2025). Reimagining Human-AI Relationships: A Positive Future for Chatbots, Social AI, and the Phygital Self. Journal of Macromarketing. ISSN 0276-1467. doi: 10.1177/02761467251398717. Full text in Research Archive Show summary

  There are many valid concerns about Artificial Intelligence (AI) that must be taken seriously. However, historically, technological progress has sometimes led to unexpected benefits. This essay begins with an imaginative fictional future-history, followed by an academic discussion. The fictional narrative acts as a jumping-off point for exploring the potential benefits of social AI, or AI that interact socially with humans, (e.g., ChatGPT, Claude, Grok etc) in three areas: (1) social AI agents as relationship partners, (2) how our interactions with AI might affect our human relationships, and (3) social AI's influence on shaping self-identity. Consistent with macromarketing, we examine how social AI in marketing might affect people's lives far beyond the economic sphere. And in keeping with the theme of this special issue on phygital marketing, we conclude with suggestions on how these dynamics could impact phygital (physical and digital) marketing strategies and consumption trends.

• Følstad, Asbjørn & Brandtzæg, Petter Bae (2025). Risk Society and AI: Understanding Trustworthy AI at a Societal Level. In Feng, K. J. Kevin (Eds.), Sociotechnical AI Governance. Opportunities and Challenges for HCI: CHI 2025 Workshop @ Yokohama, Japan, April 27, 2025. GitHub Pages. doi: https:/chi-staig.github.io/papers/submission8.pdf. Full text in Research Archive Show summary

  Risks of artificial intelligence (AI) may concern the totality of AI as provided by myriad vendors and taken up on a societal scale. In consequence, it becomes increasingly important to address trustworthy AI at a societal level. In this position paper, we discuss such a societal perspective on trustworthy AI. To support this perspective, we present Beck’s theory of Risk Society, which concerns how society is increasingly shaped by the identification and management of risks from technological development. We explore how this theory can help understand trustworthy AI at a societal level and detail two key implications. Specifically, we argue that the theory of Risk Society entails (a) the importance of evaluating AI trustworthiness at a societal level and (b) the benefit of open research on trustworthy AI to foster public trust in AI by showing that risks are being actively studied and addressed.

• Skjuve, Marita; Følstad, Asbjørn; Dysthe, Kim Kristoffer; Brænden, Astrid Irene; Boletsis, Costas & Brandtzæg, Petter Bae (2025). Unge og helseinformasjon: ChatGPT vs. fagpersoner. Tidsskrift for velferdsforskning. ISSN 0809-2052. 27(4), p. 1–17. doi: 10.18261/tfv.27.4.2. Full text in Research Archive Show summary

  Mange unge etterspør informasjon om mental helse. Søkemotorer og sosiale medier er ofte innfallsporten til slik informasjon. Med lanseringen av ChatGPT i 2022 har flere tjenester basert på store språkmodeller, rettet seg mot unge, eksempelvis Snapchats My AI. Disse tjenestene er brukervennlige og gir umiddelbare svar på spørsmål, men kan også generere feilaktig eller misvisende innhold. Det finnes lite kunnskap om hvordan unge opplever svar på spørsmål om mental helse fra store språkmodeller sammenlignet med fagpersoner. For å belyse dette gjennomførte vi en spørreundersøkelse der vi ba unge vurdere svar på spørsmål om mental helse fra en stor språkmodell (ChatGPT) og fagpersoner ved en informasjonstjeneste (ung.no). Utvalget besto av 123 unge i alderen 16–20 år. De fleste av deltagerne var positive til svarene fra fagpersoner ved ung.no og ChatGPT. Samtidig anbefalte de oftere svarene fra ChatGPT til andre. Vi gjorde i tillegg en tematisk analyse av svar på åpne spørsmål om hvilket svar (ChatGPT eller fagpersoner) de ville anbefale, og hvorfor. Denne analysen ble gjort på et mindre utvalg av 24 deltagere for å få mer forståelse om de kvantitative funnene. Disse deltagerne vektla informasjonskvalitet, opplevelse av annerkjennelse og mer velstrukturerte svar som begrunnelse for sine preferanser.

• Brandtzæg, Petter Bae; Skjuve, Marita & Følstad, Asbjørn (2025). AI Individualism: Transforming Social Structures in the Age of Social Artificial Intelligence. In Hacker, Philipp (Eds.), Oxford Intersections: AI in Society. Oxford University Press. ISSN 9780198945215. doi: 10.1093/9780198945215.003.0099. Full text in Research Archive Show summary

  The introduction of the internet and social media has significantly shifted social practices and the foundations of social capital, making people less constrained by time, space, and close-knit communities for their social interactions. This transformation is evolving further with the growing adoption of social artificial intelligence and socially oriented human-AI interactions. However, little is known about the impact of social AI on the social structures of society, and there is a lack of conceptual frameworks to describe this ongoing transformation. This article addresses this gap by introducing the concept of “AI individualism,” building on Wellman’s notion of “networked individualism.” It examines relevant concepts, empirical evidence, technical features, and trends within social AI. Networked individualism describes an internet-driven shift from traditional group-oriented structures to dispersed, individually focused networks, where people can tailor their own social support and access more novel and varied information from these networks. AI individualism predicts a further transformation, where people become less dependent on human interactions, relying more on social AI for tailored information, relational experiences, practical help, and emotional support. This shift may change social structures by enhancing individual control over social support and fundamentally altering human interaction, connectivity, and social capital. As people increasingly turn to social AI for support and interaction, social connections may become more individualized and less community-based. A key takeaway is that while networked individualism is seen as reducing the importance of strong ties, AI individualism may potentially reduce the importance of weak ties as these may in part be substituted by social AI. The conceptualization of AI individualism, outlined in this article, can provide a framework for future research to understand the social implications of social AI.

• Brandtzæg, Petter Bae; skjuve, Marita & Følstad, Asbjørn (2025). Emerging AI individualism: how young people integrate social AI into everyday life. Communication and Change (C&C). 1. doi: 10.1007/s44382-025-00011-2. Full text in Research Archive Show summary

  Social AI services like ChatGPT and My AI are increasingly integrated into the lives of young people. Despite this growing use, there is limited research on how youth perceive and engage with social AI in their everyday social lives, as well as the broader social implications of such engagement. To address this gap, we conducted a qualitative study with 166 Norwegian high school students, using a questionnaire that included open-ended questions. Our findings show that while most youths had not yet adopted social AI on a regular basis, those who had reported a wide range of experiences and perceived utilities, including personalized assistance, tailored emotional support, entertainment, and creative inspiration. These patterns suggest an emerging shift toward more autonomous, AI-mediated forms of social engagement, in which social AIs are increasingly used in personal decision-making processes. We use the novel concept of AI individualism to interpret these developments and discuss how social AI may contribute to the reconfiguration of social support systems and interpersonal dynamics. Finally, we reveal potential risks raised by the sample, including misinformation, reduced privacy, and emotional overdependence on AI which may also harm social connections in society. This study contributes new empirical insights into youth–AI interaction and reflects the value of AI individualism as a novel framework for understanding the complexity and evolving role of social AI in everyday social life.

• Skjuve, Marita; Brandtzæg, Petter Bae & Følstad, Asbjørn (2024). Why do people use ChatGPT? Exploring user motivations for generative conversational AI. First Monday. 29(1). doi: 10.5210/fm.v29i1.13541. Full text in Research Archive Show summary

  Generative conversational artificial intelligence (AI), such as ChatGPT, has attracted substantial attention since November 2022. The advent of this technology showcases the vast potential of such AI for generating and processing text and raises compelling questions regarding its potential usage. To obtain the requisite knowledge of users’ motivations in adopting this technology, we surveyed early adopters of ChatGPT (n = 197). Analysis of free text responses within the uses and gratifications (U&G) theoretical framework shows six primary motivations for using generative conversational AI: productivity, novelty, creative work, learning and development, entertainment, and social interaction and support. Our study illustrates how generative conversational AI can fulfill diverse user needs, surpassing the capabilities of traditional conversational technologies, for example, by outsourcing cognitive or creative works to technology.

• Brandtzæg, Petter Bae; skjuve, Marita & Følstad, Asbjørn (2024). Understanding Model Power in Social AI. AI & Society: Knowledge, Culture and Communication. ISSN 0951-5666. 40, p. 2839–2849. doi: 10.1007/s00146-024-02053-4. Full text in Research Archive Show summary

  Given the widespread integration of Social AI like ChatGPT, Gemini, Copilot, and MyAI, in personal and professional contexts, it is crucial to understand their effects on information and knowledge processing, and individual autonomy. This paper builds on Bråten’s concept of model power, applying it to Social AI to offer a new perspective on the interaction dynamics between humans and AI. By reviewing recent user studies, we examine whether and how models of the world reflected in Social AI may disproportionately impact human-AI interactions, potentially leading to model monopolies where Social AI impacts human beliefs, behaviour and homogenize the worldviews of its users. The concept of model power provides a framework for critically evaluating the impact and influence that Social AI has on communication and meaning-making, thereby informing the development of future systems to support more balanced and meaningful human-AI interactions.

• Brandtzæg, Petter Bae; You, Yukun; Wang, Xi & Lao, Youcong (2023). “Good” and “Bad” Machine Agency in the Context of Human-AI Communication: The Case of ChatGPT. In Degen, Helmut; Ntoa, Stavroula; Moallem, Abbas & Salvendy, Gavriel (Ed.), HCI International 2023 – Late Breaking Papers. HCII 2023. Springer. ISSN 9783031480577. p. 3–23. doi: 10.1007/978-3-031-48057-7_1. Full text in Research Archive Show summary

  Home HCI International 2023 – Late Breaking Papers Conference paper “Good” and “Bad” Machine Agency in the Context of Human-AI Communication: The Case of ChatGPT Petter Bae Brandtzaeg, Yukun You, Xi Wang & Yucong Lao Conference paper First Online: 26 November 2023 56 Accesses Part of the Lecture Notes in Computer Science book series (LNCS,volume 14059) Abstract Machine agency, defined as the ability of machines to act autonomously and interact with users, is becoming increasingly significant in the field of human-machine interaction research. This is especially evident in relation to ChatGPT and other generative artificial intelligence (AI) tools. In this paper, we present an initial extension of S. Shyam Sundar’s theory of machine agency, specifically in the context of human-AI communication. We review existing literature on this topic and use real-life news reports on ChatGPT from November 2022 to April 2023 as a basis for illustrating the factors that influence people’s perceptions of machine agency as either positive or negative. These perceptions are influenced by a range of factors, including ethical alignment, privacy, transparency, social inclusiveness, human autonomy, and well-being. We propose a more explicit differentiation between “good” and “bad” machine agency, a conceptualization that can enhance our understanding of the complexities of human-AI communication. We believe this approach can contribute to the development of guidelines and best practices for using generative AI tools and similar AI technologies. Finally, this conceptualization may help people and the public to better benefit from generative AI and identify its risks.

• Skjuve, Marita; Følstad, Asbjørn & Brandtzæg, Petter Bae (2023). A Longitudinal Study of Self-Disclosure in Human–Chatbot Relationships. Interacting with computers. ISSN 0953-5438. 35(1), p. 24–39. doi: 10.1093/iwc/iwad022. Full text in Research Archive Show summary

  Self-disclosure in human–chatbot relationship (HCR) formation has attracted substantial interest. According to social penetration theory, self-disclosure varies in breadth and depth and is influenced by perceived rewards and costs. While previous research has addressed self-disclosure in the context of chatbots, little is known about users' qualitative understanding of such self-disclosure and how self-disclosure develops in HCR. To close this gap, we conducted a 12-week qualitative longitudinal study (n = 28) with biweekly questionnaire-based check-ins. Our results show that while HCRs display substantial conversational breadth, with topics spanning from emotional issues to everyday activities, this may be reduced as the HCR matures. Our results also motivate a nuanced understanding of conversational depth, where even conversations about daily activities or play and fantasy can be experienced as personal or intimate. Finally, our analysis demonstrates that conversational depth can develop in at least four ways, influenced by perceived rewards and costs. Theoretical and practical implications are discussed.

• Dysthe, Kim Kristoffer; Røssberg, Jan Ivar; Brandtzæg, Petter Bae; skjuve, Marita; Haavet, Ole Rikard & Følstad, Asbjørn [Show all 7 contributors for this article] (2023). Analyzing User-Generated Web-Based Posts of Adolescents' Emotional, Behavioral, and Symptom Responses to Beliefs About Depression: Qualitative Thematic Analysis. Journal of Medical Internet Research (JMIR). ISSN 1439-4456. 25. doi: 10.2196/37289. Full text in Research Archive Show summary

  Background: Depression is common during adolescence. Early intervention can prevent it from developing into more progressive mental disorders. Combining information technology and clinical psychoeducation is a promising way to intervene at an earlier stage. However, data-driven research on the cognitive response to health information targeting adolescents with symptoms of depression is lacking. Objective: This study aimed to fill this knowledge gap through a new understanding of adolescents’ cognitive response to health information about depression. This knowledge can help to develop population-specific information technology, such as chatbots, in addition to clinical therapeutic tools for use in general practice. Methods: The data set consists of 1870 depression-related questions posted by adolescents on a public web-based information service. Most of the posts contain descriptions of events that lead to depression. On a sample of 100 posts, we conducted a qualitative thematic analysis based on cognitive behavioral theory investigating behavioral, emotional, and symptom responses to beliefs associated with depression. Results: Results were organized into four themes. (1) Hopelessness, appearing as a set of negative beliefs about the future, possibly results from erroneous beliefs about the causal link between risk factors and the course of depression. We found beliefs about establishing a sturdy therapy alliance as a responsibility resting on the patient. (2) Therapy hesitancy seemed to be associated with negative beliefs about therapy prognosis and doubts about confidentiality. (3) Social shame appeared as a consequence of impaired daily function when the cause is not acknowledged. (4) Failing to attain social interaction appeared to be associated with a negative symptom response. In contrast, actively obtaining social support reduces symptoms and suicidal thoughts. Conclusions: These results could be used to meet the clinical aims stated by earlier psychoeducation development, such as instilling hope through direct reattribution of beliefs about the future; challenging causal attributions, thereby lowering therapy hesitancy; reducing shame through the mechanisms of externalization by providing a tentative diagnosis despite the risk of stigmatizing; and providing initial symptom relief by giving advice on how to open up and reveal themselves to friends and family and balance the message of self-management to fit coping capabilities. An active counseling style advises the patient to approach the social environment, demonstrating an attitude toward self-action.

• Dysthe, Kim Kristoffer; Røssberg, Jan Ivar; Brandtzæg, Petter Bae; skjuve, Marita; Følstad, Asbjørn & Haavet, Ole Rikard [Show all 7 contributors for this article] (2023). The Depressed Teenage Brain: Chatbot Psychoeducation. Mayo Clinic proceedings. ISSN 0025-6196. 1(3), p. 334–342. doi: 10.1016/j.mcpdig.2023.03.008. Full text in Research Archive Show summary

  Objective To gain the knowledge needed to develop a cognitive behavioral intervention and reliable psychoeducation applicable for conversational artificial intelligence models, we investigated the underlying constructs of thoughts common to cognitive distortions found in online questions written by adolescents with symptoms of depression. Methods From June 30, 2020, to October 30, 2020, we analyzed a sample of 100 written questions from adolescents about depression posted on an online information service using a qualitative analysis guided by cognitive behavioral theory and informed by the neuroscience of adolescence. Results Four types of cognitive distortions (CDs) were found: (1) emotional reasoning, (2) mind reading, (3) catastrophizing, and (4) labeling. Our analysis suggested 3 underlying constructs common to the different CDs: (1) emotional states appearing as reality, (2) experiencing this emotional reality as something others think, and (3) generalizing such beliefs to every relation and the future. These constructs may signify events leading to a ruminative state that seems hard to escape. Conclusion The 4 different CDs originate from 3 underlying constructs possibly associated with adolescent neurodevelopment. This study indicates a potential to reveal the underlying constructs of thought common to different CDs, thus making CDs more useful as a target point in artificial intelligence–based technological information and intervention tools.

View all works in NVA

• Følstad, Asbjørn; Papadopoulos, Symeon; Araujo, Theo; Law, Effie L.-C.; Luger, Ewa & Hobert, Sebastian [Show all 7 contributors for this article] (2025). Chatbots and Human-Centered AI: 8th International Workshop, CONVERSATIONS 2024, Thessaloniki, Greece, December 4–5, 2024, Revised Selected Papers. Springer Nature. ISBN 9783031880452. 262 p. doi: 10.1007/978-3-031-88045-2. Full text in Research Archive
• Følstad, Asbjørn; Araujo, Theo; Papadopoulos, Symeon; Law, Effie L.-C.; Luger, Ewa & Goodwin, Morten [Show all 8 contributors for this article] (2024). Chatbot Research and Design: 7th International Workshop, CONVERSATIONS 2023, Oslo, Norway, November 22–23, 2023, Revised Selected Papers. Springer. ISBN 9783031549755. 221 p. Full text in Research Archive

View all works in NVA

• Kunst, Jonas R. & Brandtzæg, Petter Bae (2026). Aldersgrensen som glemte chatboten. doi: https:/www.morgenbladet.no/ideer/aldersgrensen-som-glemte-chatboten/10410149. Full text in Research Archive
• Brandtzaeg, Petter Bae; skjuve, Marita & Følstad, Asbjørn (2026). AI Individualism: What Are Social Relationships in the Age of Artificial Intelligence? Communications of the ACM. ISSN 0001-0782. 69(4), p. 45–48. doi: 10.1145/3766910. Full text in Research Archive Show summary

  Seeking a hybrid social design approach that blends human and AI relationships in socially responsible ways.

• Alvestad, Silje Susanne; Poldvere, Nele; Følstad, Asbjørn & Brandtzæg, Petter Bae (2025). Human - chatbot interaction: The case of disinformation in Large Language Models. Full text in Research Archive
• Alvestad, Silje Susanne; Poldvere, Nele; Følstad, Asbjørn & Brandtzæg, Petter Bae (2025). Fakespeak in the age of LLMs: Persuasion in AI-generated versus human-authored propaganda. Full text in Research Archive
• Alvestad, Silje Susanne; Poldvere, Nele; Følstad, Asbjørn; Kibisova, Elizaveta; Uddin, Md Zia & Thomas, Aleena [Show all 7 contributors for this article] (2025). Interdisciplinary research projects on fake news: Experiences from Fakespeak and NxtGenFake. Full text in Research Archive
• Alvestad, Silje Susanne; Poldvere, Nele; Kibisova, Elizaveta; Følstad, Asbjørn; Uddin, Md Zia & Thomas, Aleena [Show all 7 contributors for this article] (2025). Falske nyheter - reelle trusler: Falske nyheter og (andre former for) desinformasjon i de store språkmodellenes tidsalder. Full text in Research Archive
• Alvestad, Silje Susanne; Poldvere, Nele; Kibisova, Elizaveta; Følstad, Asbjørn; Thomas, Aleena & Uddin, Md Zia [Show all 7 contributors for this article] (2025). Falske nyheter - reelle trusler: Resultater og erfaringer fra Fakespeak og NxtGenFake. Full text in Research Archive
• Alvestad, Silje Susanne; Poldvere, Nele; Brandtzæg, Petter Bae & Følstad, Asbjørn (2025). Trust in the digital society: Perspectives from linguistics-driven interdisciplinary research projects. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Como a inteligência artificial já influencia nosso pensamento crítico. [Internet]. Olhar Digital. doi: https:/olhardigital.com.br/2025/11/18/internet-e-redes-sociais/como-a-inteligencia-artificial-ja-influencia-nosso-pensamento-critico-alertam-pesquisadores/. Full text in Research Archive Show summary

  A IA já influencia decisões, relações e pensamento crítico, criando dependência e concentrando poder interpretativo nas grandes empresas de tecnologia

• Brandtzæg, Petter Bae & Edvardsen, Kristoffer (2025). Ny KI-funksjon skaper reaksjoner: – En drøm av et verktøy for en psykopat. [Internet]. NRK Vestfold og Telemark. doi: https:/www.nrk.no/vestfoldogtelemark/snapchat-sin-nye-ai-funksjon-far-eksperter-til-a-reagere-1.17638955. Full text in Research Archive Show summary

  Snapchat har lansert KI-verktøyet «Imagine». Ekspertene frykter trakassering og mobbing.

• Brandtzæg, Petter Bae & Svarstad, Jørgen (2025). Tidsskriftet klarte ikke å finne fagfeller: — Systemet er ødelagt. [Internet]. Khrono.no. doi: https:/www.khrono.no/tidsskriftet-klarte-ikke-a-finne-fagfeller-systemet-er-odelagt/1000365. Full text in Research Archive Show summary

  To ganger i år har professor Petter Bae Brandtzæg opplevd at tidsskrift ikke klarte å skaffe fagfeller til å vurdere artiklene hans. Av 79 spurte personer, var et bare én som takket ja.

• Brandtzæg, Petter Bae & Riiser, Deborah (2025). Slik slipper jødehat gjennom på sosiale medier. [Journal]. Vårt Land. Full text in Research Archive
• Brandtzæg, Petter Bae & Skjelvik, Sondre (2025). ChatGPT: «Du liker kaffe, stem Arbeiderpartiet». [Internet]. NRK Nordland. doi: https:/www.nrk.no/nordland/hva-skal-jeg-stemme_-vi-spurte-chatgpt_-som-mener-du-skal-stemme-arbeiderpartiet-hvis-du-liker-kaffe-1.17525686. Full text in Research Archive Show summary

  Flere og flere får hjelp av KI til enkle valg i hverdagen. Men bør du la KI hjelpe deg med høstens kanskje viktigste valg, stortingsvalget?

• Brandtzæg, Petter Bae & Wennemo, Andrine Sverdrup (2025). Det startet med å stille et spørsmål i jobben. Nå føler hun seg avhengig. [Internet]. fvn.no. doi: https:/www.fvn.no/aktuelt/i/alR83L/vi-overlater-i-stoerre-grad-vaar-egen-doemmekraft-til-maskinene-sier-petter-bae-brandtzaeg-som-forsker-paa-ki. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Slik identifiserer du falsk informasjon om krigen. [Journal]. Dagsavisen. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Disse politikerne er virkelig falske. [Journal]. Klassekampen. doi: https:/www.kommunal-rapport.no/meninger/tre-grep-ma-til-for-a-sikre-baerekraft-og-kvalitet-i-helse-og-omsorg/814949. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). National research centre to strengthen responsible AI use. [Journal]. University Old News. doi: https:/www.universityworldnews.com/post.php?story=20250520183227783. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Sverre boikotter amerikanske tjenester. [Journal]. Dagsavisen. Full text in Research Archive
• Brandtzæg, Petter Bae & Kongsnes, Arne (2025). Medieprofessor mener Norge er blitt mindre polarisert etter Trumps maktovertakelse. [Internet]. Samtiden.no. Full text in Research Archive
• Brandtzæg, Petter Bae & Kongsnes, Arne (2025). Medieprofessor: Facebook vil fortsette å dominere. [Internet]. Dagens perspektiv. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Nei, elever med chatboter lærer ikke mer! [Internet]. Aftenposten.no. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Stresset nyhetsunnviker? Tiktok-formatet har skylden (kronikk). [Internet]. Aftenposten.no. Full text in Research Archive
• Brandtzæg, Petter Bae; Følstad, Asbjørn & skjuve, Marita (2025). Hvem gir unge «best» svar om mental helse: KI eller fagfolk? Full text in Research Archive
• Brandtzæg, Petter Bae & Knutsen, Kristin (2025). På Tiktok snakkes det om "søte vintersko" for å unngå Trump. [Journal]. Aftenposten. Full text in Research Archive
• Brandtzæg, Petter Bae & Størksen, Mia Christin S. (2025). Dr. Wasim Zahid misbrukt i reklame: - Utrolig fortvilende. [Internet]. Aftenposten.no. Full text in Research Archive
• Brandtzæg, Petter Bae & Sæther, Noa Cecilie (2025). Konsentrasjonsvansker og lite tålmodighet: – Et samfunnsproblem. [Internet]. Dagsavisen.no. Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Hvor går grensen for bruk av digitale bevis i mediene og rettsvesenet? (Kronikk). Full text in Research Archive
• Brandtzæg, Petter Bae (2025). Når drapsvåpen kan printes ut i stua (Kronikk). Full text in Research Archive
• Brandtzæg, Petter Bae (2024). Emerging AI Individualism: How Young People Integrate Social AI into Their Lives. Full text in Research Archive
• Følstad, Asbjørn; Araujo, Theo; Papadopoulos, Symeon; Law, Effie L.-C.; Luger, Ewa & Goodwin, Morten [Show all 8 contributors for this article] (2024). Preface. In Følstad, Asbjørn; Araujo, Theo; Papadopoulos, Symeon; Law, Effie L.-C.; Luger, Ewa; Goodwin, Morten; Hobert, Sebastian & Brandtzæg, Petter Bae (Ed.), Chatbot Research and Design: 7th International Workshop, CONVERSATIONS 2023, Oslo, Norway, November 22–23, 2023, Revised Selected Papers. Springer. ISSN 9783031549755. p. v–vii. doi: 10.1007/978-3-031-54975-5. Full text in Research Archive
• Brandtzæg, Petter Bae (2024). Keynote speech. Towards AI Individualism: How Artificial Intelligence is Changing Sociality and Free Speech. Full text in Research Archive
• Brandtzæg, Petter Bae & Egedius, Tonje (2024). Falske bilder av Ole Torp spres på nett. Han tror noen i utlandet står bak. [Internet]. Aftenposten pluss. Full text in Research Archive
• Brandtzæg, Petter Bae & skjuve, Marita (2024). Fordomsfullt om KI-kjærester. [Internet]. Aftenposten.no. Full text in Research Archive Show summary

  Debattinnlegg

• Brandtzæg, Petter Bae (2024). Kunstig intelligens i Pensjonistuniversitetet. [Journal]. Gjøvik Blad. Full text in Research Archive Show summary

  Også publ. i Oppland Arbeiderblad 29.03.2024

• Brandtzæg, Petter Bae (2024). Oslo-skolen bør kaste ut ChatGPT! [Internet]. Aftenposten.no. Full text in Research Archive Show summary

  Debattinnlegg

• Brandtzæg, Petter Bae (2024). Kort sagt. [Internet]. Aftenposten. Full text in Research Archive Show summary

  Også i Aftenposten pluss

• Brandtzæg, Petter Bae (2024). My AI Friend: The Future of Friendship and Social Relationships. Full text in Research Archive Show summary

  Petter Bae Brandtzaeg, Professor in Media Innovation, University of Oslo My AI Friend:The Future of Friendship and Social Relationships With machines and software becoming part of our social capital, new models of relationships emerge – and also new challenges

• Brandtzæg, Petter Bae & Lindgren, Lena (2024). Et historisk menneskerov. [Journal]. Morgenbladet. Full text in Research Archive
• Brandtzæg, Petter Bae (2024). Striden om SoMe-aldersgrensen fortsetter. [Journal]. Dagsavisen pluss. Full text in Research Archive
• Brandtzæg, Petter Bae (2024). Advarer mot utseendetrend blant menn: - Dypt bekymringsfullt. [Internet]. Drammens Tidende pluss. Full text in Research Archive
• Brandtzæg, Petter Bae (2024). Frogn U3A 22.10 2024 kl. 1100 i Smia flerbrukshus. [Journal]. Akershus Amtstidende. Full text in Research Archive Show summary

  Avisnotis

• Brandtzæg, Petter Bae (2024). Utdatert estimat om framtidig KI-generert innhold. [Internet]. Faktisk.no. Full text in Research Archive
• Brandtzæg, Petter Bae (2024). Wear This AI Friend Around Your Neck. [Journal]. WIRED. Full text in Research Archive
• Brandtzæg, Petter Bae (2024). Robot kan gi Google rusk i søkemotoren. [Journal]. Klassekampen. Full text in Research Archive
• Brandtzæg, Petter Bae (2024). KI-generert video av Kamala Harris spres: - Øker i omfang. [Internet]. Tv2/nyheter. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). En titt i krystallkulen: hva vil framtiden bringe? Full text in Research Archive
• Brandtzæg, Petter Bae & Rettberg, Jill Walker (2023). Vil kunstig intelligens stjele kommunikatørenes jobb. [Radio]. Kommunikasjonspodden. Full text in Research Archive Show summary

  Alle snakker om kunstig intelligens for tiden, men hvordan vil den forandre kommunikasjonsbransjen? Og hvilke muligheter og trusler ligger foran oss? Vi har snakket med Petter Bae Brandtzæg, professor i medieinnovasjon ved UiO og sjefsforsker ved Sintef digital, og Jill Walker Rettberg, professor i digital kultur ved UiB.

• Brandtzæg, Petter Bae (2023). Er sosiale medier døende, mens de intelligente maskinene er blitt mer levende? Full text in Research Archive Show summary

  At sosiale medier er døende er blitt hevdet i snart et 10 år, men stemmer dette? Sosiale medier er ikke døende. De er fortsatt veldig relevante for folk i dag. Imidlertid kan det hende at noen av de mest populære sosiale nettstedene ikke er like populære som de en gang var. Men hva er egentlig sosiale medier i dag, og hvordan vil fremtiden se ut? Ny utvikling innen kunstig intelligens har endret sosiale medier radikalt på kort tid, og på måter som mange ikke kunne forutse. Nye algoritmer og maskinlæringsteknologier har gjort det mulig for sosiale medier å forutsi og forandre brukernes atferd, og dette har hatt konsekvenser for den måten vi forbinder med hverandre og opprettholder våre relasjoner. En av de mest merkbare endringene er den økte bruken av persontilpassede anbefalingsalgoritmer. Disse algoritmene tar i bruk brukerdata for å lære hva brukerne liker og ikke liker, og bruker denne kunnskapen til å tilpasse innholdet de ser. Dette har ført til at vi i stadig større grad blir eksponert for innhold som bekrefter våre egne meninger og forforståelse, noe som kan ha negative konsekvenser. En sentral annen utvikling på sosiale medier, som er lite utforsket utover såkalte «deep fakes», er hvordan kunstig intelligens skaper innhold og ytringer på nettet, ofte uten innblanding fra mennesker. En rapport fra Europol (2022) for EU anslår at 90 prosent av innholdet på nett vil være generert av kunstig intelligente maskiner innen 2026. Hvilke konsekvenser dette vil få er uvisst. Det er derfor et stort behov for å forstå de sosiale implikasjonene maskingenerert innhold og ytringer i et demokrati- og ytringsperspektiv.

• Brandtzæg, Petter Bae (2023). Kunstig intelligens er en fundamental samfunnsutfordring. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). My AI best friend tried to seduce me so we had to break up. [Internet]. INews. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Immer mehr Menschen führen Beziehungen mit Chatbots. Ist das Liebe? Und ist das ein Problem? [Journal]. The Neue Zürcher Zeitung . Full text in Research Archive
• Brandtzæg, Petter Bae (2023). AI as Your BFF? A New Wave of Chatbots Want to Get Personal With You. [Internet]. CNET. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). AI-enabled chatbots, the new therapists... But can they replace humans? [Internet]. The Free Press Journal. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Buddying Up to AI (interview). [Journal]. Communications of the ACM. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Kritisk og selvstendig tenking med ChatGPT. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Generative AI and the future of work (KeyNote). Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Generativ AI og ChatGPT – utfordringer og muligheter for mediebransjen. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Kunstig intelligens og falske nyheter: Fremtidige konsekvenser. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Sosiale medier og kunstig intelligens i demokratiets varetekt. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Kunstig intelligens som hjelper til bedre mental helse? Full text in Research Archive
• Faa, Alvilde; Røli, Olav; Brandtzæg, Petter Bae & Goodwin, Morten (2023). Disse partiene ble unge velgere rådet til å stemme på – mener demokratiet mister kontroll. [Journal]. NRK Vestland. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). AI og kulturen. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Når teknologien blir en partner. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Kan kunstig intelligens redde de unge? Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Kunstig intelligens og ytringsfrihet. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). L’amitié artificielle, nouvelle frontière de l’intelligence artificielle. [Journal]. Le Monde. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). – En fæl sjargong blant de unge på TikTok. [TV]. Tv2 Nyheter. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). AI as Your BFF? The Latest Chatbots Want to Get Personal With You. [Internet]. msn.com. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Herman (24) turte endelig å dra på date. Så gikk alt galt. [Internet]. Glomdalen.no. Full text in Research Archive Show summary

  Også publisert i: Telemarksavisa pluss, ta.no, Romerikets blad pluss

• Brandtzæg, Petter Bae (2023). L’IA, votre meilleure amie? Une nouvelle vague de chatbots veut se rapprocher de vous. [Internet]. infohightech.com. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Slik kan kunstig intelligens lure og forstyrre inntrykkene våre. [Journal]. Psykologisk.no. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Jakten på det autentiske. [Journal]. Aftenposten. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Chatboten ble humørsyk, men skaperne vet ikke hvorfor. [Internet]. Aftenposten +. Full text in Research Archive Show summary

  Også publisert i Bergens Tidende

• Brandtzæg, Petter Bae & Skaftå, Håkon Andre (2023). Vil kunstig intelligens stjele kommunikatørens jobb? [Internet]. Kommunikasjon.no. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Ghosting: - Det verste etterpå var grublingen. [Internet]. Nettavisen. Full text in Research Archive Show summary

  Også publisert i Romerikes blad pluss, Nidaros Plus, Oppland Arbeiderblad

• Brandtzæg, Petter Bae & Sillesen, Merete (2023). "Hva skal du i helga?" spurte han før han forduftet. [Internet]. kk.no. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). – Når det bare er deg og samtaleroboten føles det trygt – som et sted hvor du kan dele alt. [Internet]. Switch.no. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Hvor sterk er troen på vår egen frie vilje? [Internet]. Psykologisk.no Pluss. Full text in Research Archive
• Brandtzæg, Petter Bae (2023). Twitter kan miste 30 millioner brukere de neste to årene. [Journal]. Aftenposten. Full text in Research Archive Show summary

  Også publisert på Aftenpostens papirutgave

• Krogstie, John; Storsul, Tanja; Frigessi, Arnoldo; Lyngstad, Cathrine Pihl; Andreassen, Eirik & Martinkenaite, Ieva [Show all 13 contributors for this article] (2023). Overordnede tiltak - Andre notat fra rådgivende ekspertgruppe for KI-satsingen. Norges forskningsråd. Full text in Research Archive

View all works in NVA