Evidence (84 claims)
Search and filter individual claims pulled from the papers. Looking for a specific finding ("what's the effect on wages?"), you're in the right place. Want to compare whole outcome categories against each other instead? Use the Evidence Explorer.
The board below groups claims two ways: by broad theme (nine paper-level topics) and by outcome category (the 34 claim-level outcomes that the Explorer and Syntheses also use).
Browse by theme
Nine broad, paper-level topics. Click one to filter the claims below.
Adoption
21267 claims
Filter claims →
Productivity
17978 claims
Filter claims →
Governance
17038 claims
Filter claims →
Human-AI Collaboration
16914 claims
Filter claims →
Org Design
11104 claims
Filter claims →
Innovation
11087 claims
Filter claims →
Labor Markets
6711 claims
Filter claims →
Skills & Training
5616 claims
Filter claims →
Inequality
4343 claims
Filter claims →
Claims by outcome category
Counts by direction of finding. These are the same 34 outcome categories the Explorer compares and the Syntheses are written for. A linked row has a published synthesis.
| Outcome | Positive | Negative | Mixed | Null | Total |
|---|---|---|---|---|---|
| Other | 1880 | 496 | 296 | 1854 | 4721 |
| Organizational Efficiency | 2906 | 665 | 438 | 180 | 4210 |
| Governance & Regulation | 2162 | 929 | 480 | 247 | 3866 |
| Technology Adoption Rate | 1533 | 545 | 278 | 210 | 2593 |
| Decision Quality | 1391 | 534 | 321 | 173 | 2429 |
| Output Quality | 1298 | 472 | 231 | 145 | 2153 |
| AI Safety & Ethics | 682 | 821 | 230 | 90 | 1837 |
| Research Productivity | 855 | 253 | 121 | 425 | 1675 |
| Firm Productivity | 1105 | 171 | 175 | 73 | 1531 |
| Task Allocation | 735 | 229 | 361 | 99 | 1433 |
| Market Structure | 457 | 461 | 251 | 47 | 1222 |
| Innovation Output | 673 | 94 | 108 | 36 | 913 |
| Task Completion Time | 499 | 118 | 43 | 38 | 702 |
| Firm Revenue | 458 | 130 | 61 | 26 | 677 |
| Skill Acquisition | 381 | 122 | 113 | 34 | 650 |
| Consumer Welfare | 316 | 176 | 115 | 39 | 648 |
| Employment Level | 223 | 143 | 177 | 53 | 600 |
| Error Rate | 246 | 282 | 44 | 19 | 594 |
| Fiscal & Macroeconomic | 283 | 142 | 78 | 52 | 562 |
| Inequality Measures | 103 | 329 | 106 | 13 | 552 |
| Worker Satisfaction | 225 | 185 | 63 | 30 | 503 |
| Automation Exposure | 158 | 155 | 72 | 37 | 426 |
| Regulatory Compliance | 186 | 126 | 35 | 14 | 362 |
| Team Performance | 193 | 56 | 51 | 24 | 326 |
| Developer Productivity | 224 | 58 | 27 | 13 | 323 |
| Wages & Compensation | 148 | 108 | 50 | 17 | 323 |
| Training Effectiveness | 218 | 44 | 21 | 27 | 313 |
| Job Displacement | 23 | 159 | 53 | 5 | 240 |
| Hiring & Recruitment | 109 | 61 | 32 | 11 | 215 |
| Skill Obsolescence | 16 | 107 | 26 | 6 | 155 |
| Creative Output | 71 | 44 | 28 | 6 | 150 |
| Social Protection | 58 | 31 | 12 | 3 | 104 |
| Labor Share of Income | 29 | 43 | 25 | 2 | 99 |
| Worker Turnover | 45 | 29 | 6 | 4 | 84 |
| Industry | — | — | — | 1 | 1 |
Employee turnover is driven by interactions among organizational, demographic, occupational, and labor-market factors rather than by isolated organizational characteristics.
Integrated analysis of nine nationally representative workforce datasets combining organizational, demographic, occupational, compensation, employment-history, and labor-market variables.
Employment history and compensation were the dominant predictor domains for employee turnover.
SHAP-based analysis of broader predictor domains constructed from the integrated national workforce datasets.
Organizational tenure, annual wage, age, employee benefits, and regional job openings were the most influential individual predictors of employee turnover according to SHAP analysis.
Explainable Artificial Intelligence analysis using SHAP applied to the machine-learning turnover models.
U.S. employee turnover from 2010 to 2025 followed three labor-market phases: post-recession recovery (2010–2019), COVID-19 disruption (2020), and post-pandemic adjustment (2021–2025).
Descriptive trend analysis of harmonized data from nine U.S. national workforce datasets covering 2010–2025.
Employees' pre-adoption cognitive and affective appraisals of AI significantly predict their attitudes toward AI adoption and their turnover intentions, while existing AI knowledge moderates the effect of anticipated negative outcomes on those appraisals.
The claim summarizes empirical findings from Chiu, Zhu, and Corbett (2021).
Institutional trust, perceived fairness of algorithmic decisions, and experienced autonomy mediate workers' behavioral responses to algorithmic management, including motivation, compliance, and exit.
Thematic synthesis of 39 empirical studies using a six-phase coding procedure; studies included surveys, interviews, and platform data.
Increased visibility-related stress and reduced authentic belonging may raise turnover and reduce job attachment, creating hidden organizational costs such as burnout and loss of firm-specific human capital.
Conceptual extrapolation from the proposed emotional and relational mechanisms; turnover and burnout are not empirically measured in the paper.
In a cited study of Nigerian financial-services firms, organizations with formalized upskilling programs experienced a 27% lower voluntary turnover rate among technical and digital staff.
The paper summarizes Adesola and Tunde (2021), described as a cross-sectional study of Nigerian financial-services firms; the sample size is not reported in the supplied text.
Intense performance pressure generated by algorithmic pacing can exhaust employees and trigger withdrawal behaviors, safety-compliance failures, and elevated turnover intentions.
The paper's literature synthesis cites Doan and Nguyen (2025) and Song et al. (2026).
Les pratiques organisationnelles non inclusives peuvent entraîner une perte de talents issus de groupes sous-représentés et accroître les coûts de rotation.
Conclusion intégrative fondée sur la littérature relative à la rétention, à la progression de carrière et aux biais organisationnels; aucune taille d’échantillon ni estimation de coût n’est rapportée.
Les normes de travail rigides, notamment concernant les horaires et la mobilité, pénalisent certains profils divers, dont les parents, les personnes en situation de handicap et les minorités.
Littérature académique sur les normes de travail, les contraintes organisationnelles et les inégalités professionnelles; les contextes nationaux et les tailles de cabinets sont hétérogènes.
Among the interviewees, 9 of 15 reported turnover intention, while 11 of 15 reported burnout or emotional exhaustion.
Thematic coding of interview transcripts, with participant counts reported in Table 2; no standardized outcome scales or inferential tests were used.
Introducing a single advertising-supported monetization option leads paid subscribers to show greater preference for downgrading.
Finding reported from four experiments with a total sample of N = 1063; paid users' downgrade responses are compared across alternative monetization choice sets.
Rapid changes to return-to-office policies without impact assessments can unintentionally redistribute opportunities and increase turnover among affected workers.
The paper identifies organizational pressure around productivity, culture, and real-estate costs and links rapid policy changes to potential distributional and retention effects; no turnover estimate is provided.
Trust mediates HR governance effects on turnover intention (indirect effect β = –0.22).
Mediation analysis in PLS-SEM on the time-lagged survey (N = 387); reported indirect effect β = -0.22.
In healthcare, professional team stability is crucial, and there is persistent nurse turnover.
Background claim grounded in prior literature cited by the paper (used to motivate the case study); not an empirical estimate from this study.
Traditional talent management practices fail to address the dynamic employee needs, which disengages employees, resulting in an increase in turnover cases.
Claim made in the paper (literature/background assertion); no specific empirical estimate or causal identification given in the excerpt.
Job dissatisfaction, lack of career growth and talent pool are also a problem in employee retention in the Indian IT industry.
Statement in the paper's introduction/background; no specific empirical test or sample size cited for this statement in the excerpt.
Relative to workers in cart facilities, workers in robotics facilities report increased turnover intentions.
Survey measures of turnover intentions compared across facility types in the >1,500-worker sample across 16 FCs.
These factors (surveillance anxiety, loss of autonomy, deskilling) negatively affect worker well-being and contribute to turnover.
Secondary data literature review of peer-reviewed research and industry evidence published 2022–2026 (method: secondary data review / synthesis). The paper synthesizes prior empirical and theoretical studies but does not report an original sample size.
Incumbent workers in more robot-exposed industries are unlikely to transition outside manufacturing over 2014-2021.
Longitudinal worker-level analysis of the 2014 manufacturing cohort through 2021 showing low rates of transition out of manufacturing for workers in higher-exposure industries (administrative employer-employee data).
Secure attachment further moderated the indirect effect of organizational AI adoption on employees' turnover intentions via identity threat (i.e., it attenuated the mediated effect).
Moderated mediation (conditional indirect effect) analysis reported on three-wave survey data of 312 employees; secure attachment reported to weaken the indirect AI adoption → identity threat → turnover intentions pathway.
The study identifies specific retention issues including rigid work practices, a predominantly masculine culture, and occurrences of bullying and harassment.
Findings from thematic analysis of 23 interviews using NVivo 13; participants' accounts raised these specific themes as retention-related issues.
Workplace stress is associated with lower employee retention.
PLS-SEM analysis on a cross-sectional survey of N = 350 pharmaceutical workers in Karnataka, India (purposive sampling). Reported direct path: Stress → Retention, β = 0.321, p < 0.001. (Note: the paper interprets this as stress reducing retention; sign/coding conventions of the variables are not detailed in the summary.)
There is sizable attrition in the pipeline from applicant admission through to direct employment of AI graduates, indicating leakages at multiple stages (application → admission → graduation → employment).
Quantification of human-resource losses across pipeline stages using the monitoring dataset for the 191 institutions; descriptive counts/percentages of entrants, admitted students, graduates, and those directly employed in AI roles (pipeline loss metrics reported in paper).
Onboarding and retention are unchanged after adoption.
Difference-in-differences estimates comparing onboarding and retention metrics between adopting projects and matched non-adopting controls; reported as no significant change post-adoption.
The study shifts retention analysis from descriptive correlations and surveys toward actionable, employee-level predictions and policy evaluation.
Combination of objective HR records (IBM dataset), predictive modeling (logistic regression), calibration, XAI tools (SHAP, LIME), and counterfactual policy simulations to evaluate intervention effects at individual and aggregate levels.
Local explainability (SHAP and LIME) can identify employee-specific intervention levers for targeted retention actions.
Use of SHAP and LIME for local explanations of individual predictions; counterfactual simulations applied at the employee level to estimate impact of feature changes on that employee's calibrated attrition probability.
Seventy-nine percent of organizations use retention and engagement analytics to detect retention risks and design targeted interventions.
Survey findings on the use of HR analytics for employee retention and engagement.
UK international businesses using predictive HR analytics report 18% better employee retention rates than organizations using traditional approaches.
The study's abstract reports this comparison in the context of its mixed-methods research, which included 102 completed questionnaires; the statistical basis and definition of the retention measure are not specified.
After AI-powered user-portrait deployment, the average 90-day customer repurchase rate among the 50 sampled enterprises increased from 6.23% to 11.46%, a reported increase of 5.23%.
Pre/post comparison of average operational indicators for 50 domestic cross-border e-commerce enterprises using H1 2025 and H1 2026 data.
Random Forest achieved 91.6% predictive accuracy and Gradient Boosting achieved 90.9% predictive accuracy in employee-turnover classification, ranking below XGBoost.
Comparative evaluation of seven classification algorithms using cross-validation and multiple performance metrics.
Among the seven evaluated machine-learning algorithms, XGBoost achieved the highest employee-turnover predictive accuracy at 92.8%.
Comparative supervised machine-learning evaluation of Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, Artificial Neural Network, Gradient Boosting Machine, and XGBoost using cross-validation and multiple performance metrics.
Workforce mobility remained above historical levels during the post-pandemic period despite recent stabilization.
The study's national labor-market trend analysis using workforce datasets covering 2010–2025.
Human capital analytics improves prediction and prevention of employee turnover.
Thematic literature review identifying turnover prediction and prevention as a recurring HCA application; the review does not report a primary sample or causal effect estimate.
Predictive HR analytics can help organizations identify turnover risks and implement targeted retention interventions before employee disengagement occurs.
The review synthesizes findings from studies on predictive HR analytics, turnover prediction models, and employee retention analytics.
Workforce analytics has a positive association with employee retention across the reviewed literature.
Narrative synthesis of 22 peer-reviewed studies published between 2021 and 2026, including studies on predictive HR analytics, employee retention analytics, and talent analytics.
Responsible-AI governance, including transparency and grievance mechanisms, is expected to reduce perceived unfairness and turnover while increasing sustainable net benefits.
Testable proposition derived from the review's discussion of responsible-AI governance; the claim is not supported by a reported causal estimate in the narrative review.
The positive employment-retention effect of AI interviews persisted through four months after hiring.
The paper reports positive effects on employment persistence at two, three, and four months in the randomized comparison.
Applicants interviewed by AI were 18% more likely to have an employment spell lasting at least one month than applicants interviewed by human recruiters.
Randomized field experiment comparing AI and human interviewer conditions; the reported effect was statistically significant at p < 0.001.
The null hypotheses (no effect of AI-based recruitment on engagement and retention) were rejected, establishing that AI-based recruitment has a strong positive effect on engagement and retention.
Paper states the null hypotheses were rejected based on statistical tests. Summary provides no details about hypotheses formulation, statistical methods, test statistics, significance levels, or sample size.
Candidates who experience transparent, fair, and meaningful recruitment are more likely to stay with the organization (higher retention).
Study summary claims this relationship based on survey data and statistical testing—paper states transparent, fair, meaningful recruitment correlates with higher retention. No sample size, retention rates, or effect sizes are provided in the summary.
Statistical testing showed a very positive relationship between AI-based hiring, candidate engagement, and employee retention.
Paper summary states that statistical testing was performed and that relationships between AI hiring, engagement, and retention were 'very positive.' No test statistics, p-values, effect sizes, or sample size are provided in the summary.
AI increases retention rates by 17.9%.
Survey data collected from 304 Europe-based firms that have adopted AI recruiting tools; percent increase appears to be a self-reported, aggregate figure from that survey.
Organizations adopting predictive analytics report up to 15% reduction in employee turnover.
Reported aggregated finding in the paper's literature review and referenced business case studies; no specific sample size, study-by-study breakdown, or statistical meta-analysis provided in the text.
The AI-powered framework improves employee retention.
Claimed in the paper's summary; authors state the framework aims to "improve retention" and note validation via unspecified case studies and empirical investigations.
Comparative analysis with conventional HR approaches highlights AI’s superior ability to personalize career development plans and detect high-risk attrition cases, ensuring timely interventions.
Reported comparative analysis results in the paper indicating AI methods outperform conventional HR approaches in personalization and high-risk attrition detection on the study dataset.
AI-powered HR solutions reduce voluntary turnover rates by 27%.
Reported empirical result from comparative analysis of AI-driven HR systems versus conventional approaches on the study dataset (stated sample: >15 organizations, 12,000 records).
AI-based talent management models can help Indian IT companies to significantly increase their retention score as HR functions are personalized and the needs of every employee are addressed.
Paper's conclusion synthesizing survey and interview findings (350 participants) indicating improved retention and personalization associated with AI tools.
72 percent of the respondents believed that AI-enhanced recruitment enhanced job-role fit, which led to less dissatisfaction and turnover.
Survey result reported in the paper (350 respondents) summarizing percent who agreed AI-enhanced recruitment improves job-role fit and reduces dissatisfaction/turnover.