EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Machine competence or human experience? A source-task congruence framework for AI-generated travel recommendations

Ethan Mercer, Olivia Bennett

Digital Enterprise Studies · 2026

DOI: 10.68217/z02nrf73 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Digital Enterprise Studies, 2026. Verfasst von Ethan Mercer und 1 weiteren Autor:innen. Die Arbeit nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Travelers can now receive advice from generative artificial intelligence alongside human experts, peers and review platforms, yet prior research reports AI aversion, AI appreciation and source parity under different conditions. This conceptual paper develops a source-task congruence framework to explain when AI-attributed travel recommendations are more or less persuasive and specifies the observations that would reject the framework. It integrates source credibility theory, research on testimony and epistemic vigilance, mind perception research, tourism authenticity scholarship and work on algorithm aversion and appreciation. The framework distinguishes epistemic credibility, defined as source expertise in the credibility literature, from experiential credibility, a source-level judgment of perceived direct acquaintance with the focal place. It proposes that source attribution affects the perceived level of each credibility currency, while recommendation task warrant determines the diagnostic weight assigned to each during trust formation. Informational tasks are proposed to weight epistemic credibility more heavily, whereas experiential tasks weight experiential credibility more heavily, so persuasion depends on the match between the credibility currency a source is perceived to hold and that required by the task. Capability-enhancing information is predicted to narrow the AI-human gap more for informational than experiential tasks, providing a discriminating test against purely ability-based explanations. The paper develops eight propositions and a six-study validation agenda with prespecified effect thresholds. It also identifies the confounding of human identity with direct visitation, specifies a four-condition design to separate them, and outlines the measurement work required before testing two central variables. Conditional on empirical support, the framework implies that source labeling may carry different credibility consequences depending on whether it signals direct visitation or generic editorial oversight. The paper defines experiential credibility as a source judgment, introduces recommendation task warrant as distinct from subjectivity and hedonicity, and specifies tests separating source-task congruence from ability, originality, anthropomorphism- and human-favoritism-based accounts. All propositions remain untested.

Zitieren

APA

Mercer, E., & Bennett, O. (2026). Machine competence or human experience? A source-task congruence framework for AI-generated travel recommendations. Digital Enterprise Studies. https://doi.org/10.68217/z02nrf73

Vancouver

Mercer E, Bennett O. Machine competence or human experience? A source-task congruence framework for AI-generated travel recommendations. Digital Enterprise Studies. 2026. doi:10.68217/z02nrf73

BibTeX
@article{Mercer2026,
  author = {Mercer, Ethan and Bennett, Olivia},
  title = {Machine competence or human experience? A source-task congruence framework for AI-generated travel recommendations},
  journal = {Digital Enterprise Studies},
  year = {2026},
  doi = {10.68217/z02nrf73}
}
Kostenlos starten →