Mechanical, analytical and intuitive AI tools reshape hotel work and improve the experience of vulnerable and minority staff, while empathetic AI has little effect on task structure; multiple combinations of AI capabilities deliver benefits when matched to task demands, underscoring the primacy of task–technology fit.

Main Finding

Mechanical, Analytical and Intuitive forms of AI significantly reshape task characteristics and improve employee experience for vulnerable and minority hotel workers; Empathetic AI does not significantly change task structure. Multiple configurations of AI capabilities can produce favourable employee outcomes, underscoring the importance of task–technology fit (TTF) when designing inclusive, human‑centred AI in service operations.

Key Points

• Theory: Uses Task–Technology Fit (TTF) to assess how AI capabilities align with service tasks.
• AI capability taxonomy:
  • Mechanical AI (routine physical/operational automation)
  • Analytical AI (data-driven decision support, prediction)
  • Intuitive AI (pattern recognition, adaptive decisioning)
  • Empathetic AI (affective sensing/response)
• Sector focus: Hospitality — four hotel functions examined: front office, housekeeping, food & beverage production, and service operations.
• Methods: Three‑phase mixed‑methods design:
• Systematic literature review (SLR) to map and classify AI applications by hotel function.
• Survey of vulnerable and minority hotel employees; Structural Equation Modelling (SEM) to test effects of AI types on task characteristics and employee experience.
• Fuzzy‑set Qualitative Comparative Analysis (fsQCA) to identify combinations of AI capabilities (configurational pathways) that lead to positive outcomes.
• Results:
  • Mechanical, Analytical, and Intuitive AI significantly transform task characteristics and enhance employee experience.
  • Empathetic AI did not show a significant effect on task structure.
  • fsQCA reveals multiple distinct configurations of AI capabilities can yield favourable outcomes—task–technology alignment matters more than any single AI capability.
• Contribution: Extends TTF theory to examine AI impacts on vulnerable and minority workers in hospitality and provides practical guidance for inclusive AI design.

Data & Methods

• Phase 1: Systematic literature review mapping AI use-cases across hotel functions (front office, housekeeping, F&B production, service).
• Phase 2: Quantitative survey of vulnerable and minority hotel employees; analysed via Structural Equation Modelling (SEM) to estimate relationships between AI capability types, task characteristics, and employee experience. (Abstract does not report sample size or survey timing.)
• Phase 3: Fuzzy‑set Qualitative Comparative Analysis (fsQCA) to uncover causal‑complex (configurational) pathways—combinations of AI capabilities—that lead to positive employee outcomes.
• Evidence synthesis: Triangulation across SLR, SEM (net effects), and fsQCA (configurational causality) to strengthen internal validity of findings.
• Limitations (inferred from abstract): Hospitality‑specific context may limit generalisability; details on sample size, representativeness, and longitudinal effects are not reported in the abstract.

Implications for AI Economics

• Labour composition and task reallocation:
  • Mechanical, Analytical and Intuitive AI function as complements to workers’ tasks in hospitality, potentially enhancing productivity and job quality for vulnerable/minority employees when well aligned with tasks.
  • Empathetic AI’s limited effect on task structure suggests lower short‑run substitutability/complementarity effects for affective AI in service tasks.
• Investment prioritisation:
  • Firms seeking productivity and inclusive outcomes should prioritise Mechanical, Analytical and Intuitive capabilities that align with specific task demands rather than pursuing empathetic AI as a default.
  • Configurational gains: value comes from combinations of AI capabilities matched to task profiles — one‑size‑fits‑all AI investments are less likely to yield optimal worker welfare or productivity gains.
• Wage, inequality and distributional concerns:
  • Positive employee experience gains imply potential for improved retention and human capital accumulation among vulnerable workers, which may mitigate some distributional harms of automation if task–technology fit is emphasised.
  • Policymakers should monitor whether productivity gains translate into equitable wage and career progression outcomes for minority and vulnerable groups.
• Policy and managerial recommendations:
  • Support training and reskilling targeted at complementarity with Analytical/Intuitive systems.
  • Design procurement and deployment strategies that explicitly consider TTF to avoid mismatches that could harm job quality.
  • Evaluate AI adoption not only on efficiency metrics but also on worker experience, inclusion, and downstream labour market effects.
• Research directions for AI economics:
  • Estimate causal effects of specific AI configurations on wages, employment, and career mobility using longitudinal or quasi‑experimental designs.
  • Quantify returns to complementary human capital investments tied to different AI types.
  • Model distributional impacts across worker subgroups to inform equitable AI policy and firm investment decisions.