English

Cognitive offloading and the speedup illusion in human-AI interaction

Computers and Society 2026-05-25 v1 Human-Computer Interaction

Abstract

Large language models (LLMs) have the potential to boost human productivity by speeding up task completion -- provided users know when to offload cognitive work to them. But we do not know if users are well-calibrated in estimating these potential time savings. We conducted a preregistered large-scale behavioral study (N = 1237) to characterize mismatches between expectations and reality, with a focus on simple cognitive tasks. While actual completion times between independent completion and AI-assisted completion did not differ, participants predicted AI to be significantly faster. The same bias was not observed when imagining help from another human participant. We identify a speedup illusion where people have accurate forecasts of independent completion times but significantly underestimate AI-assisted times. Additionally, time and effort dissociate: participants reported lower subjective effort with AI despite equivalent completion times. This suggests that completion time itself is not sufficient to characterize efficiency gains.

Keywords

Cite

@article{arxiv.2605.23177,
  title  = {Cognitive offloading and the speedup illusion in human-AI interaction},
  author = {Sunny Yu and Myra Cheng and Ahmad Jabbar and Ilia Sucholutsky and Katherine M. Collins and Dan Jurafsky and Robert D. Hawkins},
  journal= {arXiv preprint arXiv:2605.23177},
  year   = {2026}
}

Comments

Proceedings of the 48th Annual Meeting of the Cognitive Science Society

R2 v1 2026-07-22T07:27:31.900Z