@inproceedings {pub6506,
	title = {Shaping Expectations of AI: A Basic Psychological Needs Perspective Across Cultures},
	author = {Christiane Attig AND Patricia Wollstadt AND Martina Hasenj{\"a}ger AND Alan  Sarkisian AND Jouh Yeong Chew AND Christiane Wiebel},
	year = {2026},
	month = {July},
	abstract = {Interactions with artificial intelligence (AI) systems, particularly LLM-based chatbots, have become an integral part of daily work and private life for millions of people worldwide in recent years. However, given the diverse purposes and embodiments that AI systems can assume, hands-on experience remains only one source shaping expectations for future AI encounters. Other influential factors include the perceived role of AI (e.g., an AI seen as a tool is expected to behave less sociably than one seen as a teammate), user goals (e.g., an AI supporting emotional needs is expected to be more empathetic than one supporting a professional task), and cultural background (e.g., whether robots are positively valued in popular culture or often framed as a threat to humanity). In this work, we propose that Basic Psychological Needs Theory (BPNT) provides a useful framework for organizing and explaining variability in people{\textquoteright}s expectations toward AI.
In an intercultural online mixed-methods study with 200 participants each from the United Kingdom and Japan, participants were instructed to complete a story stem in which they imagined interacting with an AI system in either a medical or a household scenario in the near future. After the qualitative writing task, they completed a set of quantitative questionnaires, including measures assessing the perceived AI role and the fulfillment of basic psychological needs. The results show that expectations toward AI systems cluster around the anticipated satisfaction of autonomy, competence, and relatedness needs, with these expectations shaped by the AI{\textquoteright}s perceived role, the user{\textquoteright}s goals, and cultural context. For instance, viewing the AI as a teammate rather than a tool was associated with higher anticipated relatedness satisfaction. Overall, the findings underscore the foundational relevance of basic psychological needs for meaningful human{\textendash}AI interaction and reveal distinct cultural patterns in how AI systems are expected to support these needs. },
	publisher = {unpublished},
	booktitle = {SDT Conference 2026},
	city = {Geneva, Switzerland}
}