Editorial summary. This is our text summary of an article. Charts, figures, and the author’s full voice are at the original — read it there .
Licence. The original is licensed under CC BY 4.0. This page adapts it: the summary and analysis are ours, not the authors’.
Editorial verdict
Methodologically cautious but limited: the study's core differentiation of AI capabilities across HRM functions is conceptually valuable, but mixed model fit, discriminant validity failures on three key construct pairs, non-probability sampling, and a cross-sectional design mean the specific path coefficients cannot be treated as reliable evidence — treat the framework as a hypothesis-generator, not a findings source.
Executive summary
This quantitative cross-sectional study examines how three distinct AI capability types — knowledge sharing systems (KSS), decision support systems (DSS), and smart systems (SS) — relate to three HRM functions (recruitment, training, and performance appraisal) in Saudi Arabian private-sector organisations. The authors argue that treating AI as a single homogeneous technology obscures function-specific relationships, and that capability-function alignment matters for HRM effectiveness. Data were collected from 431 employees via structured electronic questionnaire and analysed using structural equation modelling (SEM). Key findings show KSS positively associated with recruitment (β = 0.784), DSS with training (β = 0.854), and SS with both training (β = 0.119) and performance appraisal (β = 0.798); five of nine hypothesised relationships were not statistically significant. However, the measurement model demonstrated mixed fit (CFI = 0.853, TLI = 0.833, below the 0.90 threshold), and discriminant validity was not uniformly established for three construct pairs. The authors acknowledge these limitations and frame the theoretical contributions — grounded in Resource-Based View and contingency theory — as interpretive rather than empirically confirmed. The study is situated within Saudi Vision 2030's digital transformation agenda.
Key insights
- 1Different AI capability types show distinct associations with specific HRM functions: KSS with recruitment, DSS with training, and SS with training and performance appraisal — suggesting that capability-function fit matters more than general AI adoption.
- 2Five of nine hypothesised AI capability–HRM function relationships were not statistically significant, reinforcing that AI does not produce uniform outcomes across all HRM activities.
- 3Measurement model limitations — including CFI and TLI below 0.90 and discriminant validity failures on three key construct pairs (KSS–recruitment, DSS–training, SS–performance appraisal) — mean the strongest reported associations are also those with the least empirically distinct constructs, warranting significant caution.
Practical takeaways
- HR practitioners evaluating AI tools may find it useful to assess which specific AI capability type (knowledge retrieval, analytical decision support, or adaptive/personalised automation) is most aligned with the HRM function in question, rather than evaluating AI as a single category.
- The study's authors note that implementation of AI-enabled HRM warrants attention to data quality, transparency, fairness, privacy, and ongoing human oversight — factors identified as potentially moderating effectiveness regardless of capability type.
Frameworks mentioned
Resource-Based View (RBV)
A theoretical lens, originating with Barney (1991), treating AI capabilities as organisational resources that are valuable, rare, inimitable, and non-substitutable, potentially supporting competitive advantage in HRM.
Contingency Theory
Used as an interpretive framework suggesting that the effectiveness of AI capabilities in HRM depends on the degree of fit between capability characteristics and the informational or task demands of specific HRM functions.
Fornell–Larcker Criterion
A discriminant validity assessment method comparing the square root of average variance extracted (AVE) for each construct against inter-construct correlations.
Heterotrait–Monotrait Ratio (HTMT)
A discriminant validity test used alongside the Fornell–Larcker criterion, with thresholds of 0.85 and 0.90 applied to evaluate construct distinctiveness.
References
- Aleisa et al. (2023).AI recruiting model for the Saudi labor market.
- Al-Faiyami (2023).Measurement items for KSS, DSS, and recruitment constructs.
- Mehran (2024).Measurement items for smart systems construct.
- Al-Ayed (2025).Measurement items for training construct.
- Alrashedi & Abbod (2020).Measurement items for performance appraisal construct.
- Tabachnick and Fidell (2014).Multivariate Analysis of Variance (reference for multicollinearity threshold).
- Podsakoff et al. (2003).Common method biases in behavioral research.
- Fornell & Larcker (1981).Evaluating structural equation models with unobservable variables and measurement error.
- Henseler et al. (2015).Using partial least squares path modeling in international advertising research.
Source & Provenance
Not specified
Omar A. Baakeel
November 17, 2026
Research Study
Middle East
Original source metadata is preserved. AI analysis is generated separately.
Like this? Get the Monday Decision Brief — free, every week.
No spam, unsubscribe anytime.
Rate this article
Want the full article? Read it at the original source — free, no paywall.
Read original article
