People overestimate how much mental effort AI assistance alleviates (offloading illusion), driven by overestimating the effort of independent completion while AI-assisted effort is accurately estimated¶

Claim · All claims
Evidence · 1 study · causal r3 · q3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Predicted effort reduction from AI (0.90 on the NASA-TLX scale) exceeds actual reduction (0.62); participants overestimated independent effort (2.66 predicted vs. 2.36 actual) but accurately estimated AI-assisted effort (1.76 vs. 1.74). → Sunny Yu 2026

Evidence¶

Sunny Yu 2026¶

Sunny Yu, Myra Cheng, Ahmad Jabbar, Ilia Sucholutsky, Katherine M. Collins, Dan Jurafsky, Robert D. Hawkins. (2026). The efficiency-gain illusion: People underestimate the rate of AI use and overestimate its benefits on simple tasks. arXiv. https://arxiv.org/abs/2605.22687 doi:10.48550/arxiv.2605.22687

q3 · i? · causal · r3

Study 2 completion/prediction samples measured subjective mental effort with the NASA-TLX scale. The article reports participants "overestimatedE H, the amount of effort required for them to complete the task independently" while AI-assisted effort estimates were accurate.

"As a result, the magnitude of effort savings is overestimated, with an average predicted reduction of 0.90 compared to an actual reduction of 0.62 (Equation (2);β=−0.28,p <0.001)."

Discussion¶

Related Claims¶

• Access to AI advice roughly doubled reported confidence in answers in one experiment — the general claim this finding bears on, with the studies behind it
• In a one-year PBL deployment, the AI-assisted pipeline was associated with instructor preparation time falling from an average of eight hours to two hours per instruction (an estimated 75% reduction). — related
• Students expect AI to help their test scores, but control students overestimate the effect relative to the experimental estimate — related
• AI assistance reduces subjective mental effort across all tasks even when it does not reduce completion time, dissociating time and effort — related
• People overestimate how much time AI assistance saves (speedup illusion), driven by miscalibration about AI-assisted completion time while independent completion time is well calibrated — related
• Prompting is a major source of AI-assisted cognitive effort: copying prompts reduces effort but not time, and verbose model responses can make AI-assisted completion slower than predicted — related
• People are well-calibrated about their own independent completion times — related
• Educators benefit from AI chatbots through time-saving administrative assistance and improved pedagogy — reports the opposite
• No differences were found on mental effort during instruction or on transfer across illusion-of-control and instruction-method conditions — related
• People use AI on simple tasks at a rate significantly higher than they predict for themselves (self-estimate miscalibration), especially on easy task variants — related
• Participants more averse to thinking are more susceptible to the speedup illusion; AI familiarity measures do not predict calibration error — related
• AI-related placebo effects lead users to overestimate their own abilities while or after being assisted by an AI — related