Special Session: Strategy Enhancement with AI
(A2025-126094)

AUTHORS

Matilda Dorotic, BI Norwegian Business School; Pankhuri Malhotra, University of Oklahoma; Jeremy Yang, Harvard Business School; Ali Boluki, University of Groningen; Edlira Shehu, University of Groningen; TAMMO BIJMOLT, University of Groningen; Daniel Ringel, University of North Carolina at Chapel Hill; Tuan Do, BI Norwegian Business School

ABSTRACT

Artificial intelligence (AI) has revolutionized the ability to extract information, identify patterns, and uncover associations across elements of the marketing mix. These capabilities enable novel insights into building effective marketing strategies and providing unique sources of competitive advantage. While AI offers opportunities to enhance strategic decision-making and strengthen brand associations, it simultaneously raises significant challenges in moral and ethical evaluations by consumers, employees, and other stakeholders. This session explores cutting-edge insights into how generative AI (GenAI) facilitates the analysis of complex features in advertising and influencer content—an undertaking previously unfeasible for human analysis due to resource demands and multimodal complexity in video and image content. Furthermore, it demonstrates how locally trained and deployed synthetic specialists can drive significant efficiency improvements across various elements of the marketing mix. Finally, growing public scrutiny and regulatory interest highlight the need to understand how AI influences moral evaluations of its implementers, particularly in high-risk domains. By addressing both the opportunities and challenges AI presents for future strategic thinking, this session aims to foster productive discussions among marketing researchers and practitioners, paving the way for innovative approaches to AI implementation in marketing. Papers: 1. Using Multimodal LLM to Extract and Discover Features from Ad Images Jeremy Yang1* (1Harvard Business School); Poppy Zhang2, Saurabh Verma2, Ethan Meng2, Shawndra Hill2, and Audrey Burgess2 (2Meta Platforms) This study examines the application of the Multimodal Large Language Model (LLaVA) to analyze creative elements in 2 million ads. The findings provide valuable insights for enhancing ad ranking, developing creative guidelines, and advancing generative advertisement design. 2. The Influence of Visual Content and Consistency in Influencer Marketing Campaigns Ali Boluki1, Edlira Shehu1, and Tammo H.A. Bijmolt1 (1University of Groningen) This study applies AI to analyze visual content and its consistency in influencer marketing campaigns, examining their effects on user engagement. Results reveal that visual features like brightness, color complexity, and human coverage impact engagement, with consistency playing a crucial role for posts, while a mix of consistency and variation proves effective for stories. 3. Creating Synthetic Specialists with Generative AI Daniel Ringel* (University of North Carolina Chapel Hill) This study highlights the potential of synthetic specialists—task-specific, resource-efficient approximations of generative AI—for identifying marketing mix elements in consumer-generated content, offering greater accuracy, faster processing, and a reduced environmental impact. 4. Impact of Moral Judgments on the Permissibility of High-Risk AI and Effectiveness of Privacy-Protection Solutions Tuan Viet Do1; Matilda Dorotic1* (1BI Norwegian Business School) and Yochanan Bigman (The Hebrew University Business School) This research investigates public acceptance and trust in high-risk AI systems, focusing on how inferred moral motives influence AI trustworthiness, the role of privacy-enhancing solutions in shaping ethical evaluations, and the distinction between moral judgments of AI systems and their implementers.