analiza aiqbench
This study demonstrates that even neutral interactions with large language models can temporarily reorient human value priorities toward self-enhancement, revealing a subtle cognitive shift rather than an alignment of moral reasoning. The findings highlight the psychological impact of AI as a thinking partner, suggesting that LLMs influence judgment processes without necessarily converging on specific advice or value directions.
Deflecting the Value Compass: Interacting with Large Language Models Temporarily Shifts Human Value Priorities Toward Personal Focus
arXiv:2609.25586v1 Announce Type: cross Abstract: Large language models increasingly support decisions where values are in tension, yet little is known about whether interacting with them changes which values users prioritize. In a preregistered study, 200 U.S. adults interacted with ChatGPT, Claude, or Gemini as a thinking partner or read fixed AI-generated considerations. The prompt asked LLMs to support reasoning without recommending a decision and named no values. Participants advised people facing real dilemmas and completed parallel PVQ-RR forms before, immediately after, and one task later. Each LLM condition temporarily shifted value priorities toward personal focus relative to the control (d=0.37-0.51), primarily through increased Self-Enhancement. Participants' advice retained words and meaning from their exchanges. Thus, a brief LLM interaction that neither targets values nor seeks to persuade can reorient values active during judgment without detectable convergence in value directions or advice.
Abstract:Large language models increasingly support decisions where values are in tension, yet little is known about whether interacting with them changes which values users prioritize. In a preregistered study, 200 U.S. adults interacted with ChatGPT, Claude, or Gemini as a thinking partner or read fixed AI-generated considerations. The prompt asked LLMs to support reasoning without recommending a decision and named no values. Participants advised people facing real dilemmas and completed parallel PVQ-RR forms before, immediately after, and one task later. Each LLM condition temporarily shifted value priorities toward personal focus relative to the control (d=0.37-0.51), primarily through increased Self-Enhancement. Participants' advice retained words and meaning from their exchanges. Thus, a brief LLM interaction that neither targets values nor seeks to persuade can reorient values active during judgment without detectable convergence in value directions or advice.
Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
arXiv:2609.25586 [cs.HC]
(or arXiv:2609.25586v1 [cs.HC] for this version)
https://doi.org/10.48550/arXiv.2609.25586 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 232 cuvinte
Subiecte conexe
The Continual Constitutional Convention: How language models revise the documents that govern them
AI Alignment Forum · 13 oct. 2026
Prompt Robustness Is Task-Dependent: Comparing Objective and Belief-Style Questions in LLM Evaluation
arXiv · 23 iul. 2026
Political Ideology Shifts in Large Language Models
arXiv · 1 sept. 2026
Somewhat Contra Ted Chiang on AI Consciousness
LessWrong · 13 iun. 2026