Generative AI and AI-generated Contents on Social Media
Permanent URI for this collection
Browse
Recent Submissions
Item Alleviating Aversion to Artificial Intelligence in Digital Advertising with Humorous Disclosures(2025-01-07) Karpinska-Krakowiak, Mags; Hajduk, Grzegorz; Mroz-Gorgon, Barbara; Skowron, LukaszArtificial Intelligence (AI) is increasingly used in creating online advertisements and posts. Yet, its prominent role in content creation remains often undisclosed due to potential aversive consumer reactions. In three studies, we examined how advertisers can transparently disclose AI authorship in ads without compromising marketing effectiveness. Study 1 was a proof-of-concept study, revealing that humor could alleviate negative evaluations (when the joke was less funny, AI received more negative ratings than a human author; however, when the joke was funnier, AI received similar ratings to a human author). Study 2 incorporated jokes into disclosures about AI authorship in advertising and found that humor reduced the negative effect of AI authorship because it shortened the psychological distance toward AI, positively impacting purchase intentions. Study 3 replicated these effects and tested a boundary condition, showing that humor mitigated algorithm aversion only when it was sufficiently funny.Item How to Craft Fake Content Efficiently: Exploring Diversity in Large Language Model Outputs(2025-01-07) Thomsen, Lars; Theuerkauf, René; Peters, RalfWith the increasing importance of language models in research, investigating specific quality metrics is crucial. This study examines the diversity of outputs generated by them under varying conditions, focusing on temperature and the number of outputs per generation. Findings indicate that higher temperatures and increased number of outputs generally result in more unique and diverse outputs, while single outputs show higher convergence at low temperatures. Lexical and semantic similarities decrease with rising temperatures, increasing diversity but potentially reducing coherence. Text length is affected by temperature and output type, creating multiple outputs producing more concise texts and single outputs generating longer texts. Multiple output crafting is also more time-efficient than single one, making it suitable for applications needing diverse, quickly generated content. Insights are applicable to various fields, including creative content generation and the production of diverse, coherent, fake content, illustrating the complex dynamics of output diversity.Item Conversational Agents Powered by Generative Artificial Intelligence Improve Understanding for Those Worried About Controversial Topics(2025-01-07) Kim, Antino; Dennis, Alan; Smith, EvelynThe combination of the partisan ideological divide on controversial societal issues and algorithmic social media curation has led to echo chambers where individuals predominantly engage in political discussions aligned with their own viewpoints. Concerns about potential conflict and judgment in discussions with those holding opposing viewpoints make people apprehensive about reaching out across the political divide, further exacerbating misunderstanding and widening the divide. Our research examines whether artificial intelligence (AI) could help bridge the ideological divide by providing the means to discuss politically charged issues without concerns about emotional conflicts and the judgment associated with human interactions. Our online experiment demonstrates that when interacting with AI (versus reading an article), individuals who have apprehensions about engaging in discussions with someone with opposing viewpoints experience significantly higher satisfaction, willingness to participate, perceived learning, and a willingness to learn more about opposing perspectives and become more tolerant towards opposite viewpoints. These findings suggest that conversations with AI chatbots may help mitigate the ideological divide, particularly for those uneasy about conversing with people holding opposing views. When participants were able to freely ask questions and converse with an AI on topics that they would feel uncomfortable discussing with a person, their attitude towards the opposing side became more positive. Overall, our study underscores the potential for AI-mediated discussions to foster greater understanding and reduce ideological divides.Item Digital "Double Hatters": Augmenting Audiovisual Creative Work with a Generative Text-to-Video Workflow(2025-01-07) Liu, Vivian; Long, Tao; Ma, Jenny; Raw, Nathan; Yang, Jiaxin; Tang, Claudia; Wang, Lulu Yueyi; Yang, Yumo; Chilton, LydiaGenerative AI is transforming work for creative professionals. Text-to-image and text-to-video tools provide alternative methods for visual content creation, circumventing the traditional workflows creative professionals have developed. This transformation raises the pressing question of whether these new generative workflows will augment or supplant creative work. To investigate this question, we introduce Generative Disco, a text-to-video system for music visualization, and use it as a technology probe for creative professionals who freelance audiovisual work. In a mixed-methods study (n=12), we observe that professionals found Generative Disco easy to use, highly expressive, and capable of supporting many professional use cases. Its generative workflow enabled professionals to become digital "double hatters", expanding their creative range into adjacent domains. We conclude on how creatives can benefit from this new means of skill mobility.Item Pandora Box or Golden Fleece: Economic Analysis of Generative AI Adoption on Creation Platforms(2025-01-07) Gao, Yi; Wang, Zhe; Huang, YanIn this paper, we employ a game-theoretic approach to analyze the impact of adopting generative AI tools on various market outcomes, including creators’ creativity level and pricing strategies, the platform’s regulatory strength, and consumer welfare. Our results show that although generative AI tools can facilitate content creation by leading to a higher quality, creators’ creativity levels and prices may experience a decline compared to the regime without such tools. We also find that even though generative AI has the potential to improve content quality, it is not always true that the utilization of such tools can lead to higher consumer welfare. Moreover, we find that the platform opts for a relatively low regulatory strength when the AI intelligence is relatively small. Interestingly, as generative AI acquires a more extensive knowledge base, the platform and creators may be less willing to adopt it.Item Scrolly2Reel: Retargeting Graphics for Social Media Using Narrative Beats(2025-01-07) Ma, Jenny; Nguyen, Duy K.; Perez, Pedro Alejandro; Chilton, LydiaContent retargeting is crucial for social media creators. Once great content is created, it is important to reach as broad an audience as possible. This is particularly important in journalism where younger audiences are shifting away from print and towards short-video platforms. Many newspapers already create rich graphics for the web that they want to be able to reuse for social media. One example is scrollytelling sequences or "scrollies" -- immersive articles with graphics like animation, charts, and 3D visualizations that appear as a user scrolls. We present a system that helps transform scrollies into social media videos. By using the scriptwriting concept of narrative beats to extract fundamental storytelling units, we can create videos that are more aligned with narration, and allow for better pacing and stylistic changes. Narrative beats are thus an important primitive to retargeting content that matches the style of a new medium while maintaining the cohesiveness of the original content.Item Introduction to the Minitrack on Generative AI and AI-generated Contents on Social Media(2025-01-07) Su, Yiran; Wang, Yichuan