Predicting the long-term impact of AI assistants on collective intelligence
- 1Seoul National University
Abstract
Artificial intelligence (AI) assistants can improve individual performance but may weaken unaided performance and limit interest diversity, raising questions about their long-term effects on collective intelligence (CI). Here, we consider an evolutionary model in which individuals predict, consult AI, and imitate better-performing peers. Classic reward structures that promote CI without AI can fail under Chatbot AI, whose predictions depend on users' interests, by reducing interest diversity or propagating bias. Directly rewarding or penalizing AI use does not reliably mitigate these effects. Instead, a balanced incentive that preserves minority expertise and rewards collective contributions can promote CI, but an additional incentive is needed to restore unaided collective accuracy. Our framework offers an explanation for emerging evidence of AI's societal impact and clarifies its long-term implications.
Related
- Collective impact of opinion formation on agent-based epidemic dynamics
- Predicting Long-Term ADAS-Cog Change From Early Response After Repetitive Transcranial Magnetic Stimulation in Alzheimer Disease
- Extracellular Vesicle Protein Cargo Reveals Potential Long-Term Biological Impact of Long COVID
- Long-term impact of the diagnosis on quality of life, social skills, and person recognition strategies in congenital prosopagnosics: A cross-sectional study
- Real-Life Performance of the Dermatology Long-Term Life Impact Measure (DLLIM) in Atopic Dermatitis: An International Study of 15,450 Patients
BibTeX
@misc{kim2026predicting,
title = {{Predicting the long-term impact of AI assistants on collective intelligence}},
author = {J. Kim and S. W. Park},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.29.755292},
url = {https://www.biorxiv.org/content/10.64898/2026.09.29.755292v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Predicting the long-term impact of AI assistants on collective intelligence
AU - J. Kim
AU - S. W. Park
PY - 2026
DA - 2026/09/29
PB - bioRxiv
DO - 10.64898/2026.09.29.755292
UR - https://www.biorxiv.org/content/10.64898/2026.09.29.755292v1
AB - Artificial intelligence (AI) assistants can improve individual performance but may weaken unaided performance and limit interest diversity, raising questions about their long-term effects on collective intelligence (CI). Here, we consider an evolutionary model in which individuals predict, consult AI, and imitate better-performing peers. Classic reward structures that promote CI without AI can fail under Chatbot AI, whose predictions depend on users' interests, by reducing interest diversity or propagating bias. Directly rewarding or penalizing AI use does not reliably mitigate these effects. Instead, a balanced incentive that preserves minority expertise and rewards collective contributions can promote CI, but an additional incentive is needed to restore unaided collective accuracy. Our framework offers an explanation for emerging evidence of AI's societal impact and clarifies its long-term implications.
ER -