68/100

Moderate evidence

ProblemPublished recordLifestyleProblem Space

When friendly AI backfires: Relationship norm misfit in chatbot service recovery

Truvace record
TRV-2026-2639raw data (JSON)
Version
v1Published
First published
Evidence type
Controlled or validated comparison
Sources
1 sourcePeer-reviewed
sha256
28633fe8307c…check it

This study identifies a context-specific boundary condition within established frameworks governing chatbot service recovery: the effectiveness of warm, relational recovery cues depends on alignment with the relationship norms operative in the exchange context. Drawing on Relationship Norms Theory and Justice Theory, we develop and test a norm-contingent inference account in which recovery effectiveness is governed by alignment between chatbot framing and the dominant relationship logic of the exchange. Across f…

The problem

In four controlled experiments on e-commerce failure scenarios, warm, communal chatbot recovery framing reduced customers' forgiveness, repurchase intention and behavioural acceptance when exchange norms were salient.

What this doesn’t fix

The findings are limited to four experiments using exchange-oriented e-commerce failure scenarios.

Sources

1 on file
  1. Computers in Human Behavior Reportsdoi:10.1016/j.chbr.2026.101409
    Peer-reviewed2026-10-28

Cite this record

v1

Truvace Impact Record TRV-2026-2639, v1: “When friendly AI backfires: Relationship norm misfit in chatbot service recovery.” Truvace, 2026-11-12. /record/TRV-2026-2639 (accessed on the day you copy it). sha256 28633fe8307c0ff0…

Change history

1 version

Every change made to this record since it was first published, and why. Nothing has changed since it went up.

  1. Publishedv128633fe8307c…

    Certified into the record

Check this record hasn’t been altered

Check this record hasn’t been altered
How to check it yourself, without trusting this page

Download any version from /api/record/TRV-2026-2639 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.