When Agents Shop for You: Role Coherence in AI-Mediated Markets
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913071395504128 |
|---|---|
| author | Alavi, Soogand Nozari, Salar |
| author_facet | Alavi, Soogand Nozari, Salar |
| contents | Consumers are increasingly delegating purchase decisions to AI agents, providing natural-language descriptions of their preferences and identity. We argue that these representations constitute an information channel, role coherence, through which sellers can infer willingness to pay without explicit disclosure by the buyer agent, leading to preference leakage. In an experiment where a language-model buyer agent shops on behalf of a verbal consumer profile, we show that seller-side inference from dialogue alone recovers willingness to pay nearly one-for-one. Comparing this setting to a numeric-budget condition with confidentiality instructions cleanly isolates role coherence as distinct from instruction-following failure. Because this leakage arises from delegation itself, it cannot be mitigated at the prompt level. Instead, we propose architectural interventions that trade off personalization against preference privacy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_26220 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | When Agents Shop for You: Role Coherence in AI-Mediated Markets Alavi, Soogand Nozari, Salar Multiagent Systems General Economics Economics Consumers are increasingly delegating purchase decisions to AI agents, providing natural-language descriptions of their preferences and identity. We argue that these representations constitute an information channel, role coherence, through which sellers can infer willingness to pay without explicit disclosure by the buyer agent, leading to preference leakage. In an experiment where a language-model buyer agent shops on behalf of a verbal consumer profile, we show that seller-side inference from dialogue alone recovers willingness to pay nearly one-for-one. Comparing this setting to a numeric-budget condition with confidentiality instructions cleanly isolates role coherence as distinct from instruction-following failure. Because this leakage arises from delegation itself, it cannot be mitigated at the prompt level. Instead, we propose architectural interventions that trade off personalization against preference privacy. |
| title | When Agents Shop for You: Role Coherence in AI-Mediated Markets |
| topic | Multiagent Systems General Economics Economics |
| url | https://arxiv.org/abs/2604.26220 |