Information Foraging With Generative AI: Usage Patterns in Germany and Israel

Abstract

Generative artificial intelligence (GenAI) alters how people seek information, regardless of its susceptibility to epistemic limitations such as producing inaccurate or biased information. Studies of individuals’ usage patterns of GenAI for accessing information remain scarce. Here, we examined how individuals perceive and use GenAI for various information purposes and at different complexity levels across cultures. Based on online surveys of representative samples from Germany (<em>N </em>= 562) and Israel (<em>N </em>= 500), the findings showed that Germans rated GenAI higher in providing comprehensive information, whereas Israelis perceived GenAI as more responsive to users’ information needs. Latent class analysis (LCA) of regular GenAI users (Germany: <em>n</em> = 159; Israel: <em>n</em> = 254) identified culturally distinct user profiles: three in Israel (e.g., Favoring Pragmatists, Reserved Experts, and Skeptical Minimalists) and four in Germany (e.g., Naïve Enthusiasts, GenAI-Savvy Abstainers, Cautious Skeptics, and Passive Optimists). Harnessing the information foraging theory, we focused on the diverging balance between the currencies (perceived benefits, i.e., responsiveness of GenAI and comprehensibility of its content), costs (epistemic AI knowledge, i.e., awareness of GenAI’s limitations), and “forager attributes” (previous experience with GenAI and knowledge of its workings). The information foraging theory prism highlighted two cross-cultural similarities: the avoidance pattern of users reporting low perceived benefits, and the inclination to utilize GenAI for more complex and risk-involving science-related information, characterizing users who demonstrated high perceived benefits and low epistemic knowledge.

Ähnliche Arbeiten