Feed Your (Information) Need

Tools

Azzopardi, Leif and Roegiest, Adam (2026) Feed Your (Information) Need. In: Search Futures @ ECIR 2026, 2026-04-02 - 2026-04-02, TUDelft.

[thumbnail of Azzopardi-2026-Feed-Your-Information-Need]
Preview
 Text. Filename: Azzopardi-2026-Feed-Your-Information-Need.pdf
Accepted Author Manuscript
License: Creative Commons Attribution 4.0 logo
Download (73kB)| Preview

Abstract

Information Retrieval (IR), as a field, was built on the assumption that information exists "out there'' and that the user's task is to find it. From ranked retrieval to Berry Picking and Information Foraging Theory, decades of research have refined how people navigate information landscapes, evaluate patches, and accumulate relevant pieces under conditions of scarcity, cost, and uncertainty. But this assumption is rapidly collapsing. With generative AI, users are no longer primarily finding information. They are growing it! Information Farming names this shift. Instead of issuing queries and traversing sources, users plant prompts, cultivate outputs through iteration, and harvest structured information within generative systems. Search effort moves from navigation to cultivation; uncertainty moves from the environment into the model; and information needs are fed through production rather than discovery. If this trajectory continues, then much of what we currently call "search'' risks becoming a legacy interaction pattern that is no longer the dominant modality. In this talk, we argue that Information Farming is evolution of Information Foraging. When information is grown rather than found, the core problems that IR research has historically addressed (i.e., ranking, coverage, query formulation, patch selection), are no longer sufficient. Feeding information needs through farming, however, introduces new challenges that demand a reorientation of the field.

ORCID iDs

Azzopardi, Leif ORCID logoORCID: https://orcid.org/0000-0002-6900-0557 and Roegiest, Adam;
CORE (COnnecting REpositories)