Evidence (14922 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
9047 claims
Filter claims →
Productivity
8066 claims
Filter claims →
Governance
7278 claims
Filter claims →
Human-AI Collaboration
6912 claims
Filter claims →
Org Design
4439 claims
Filter claims →
Innovation
4359 claims
Filter claims →
Labor Markets
3652 claims
Filter claims →
Skills & Training
3018 claims
Filter claims →
Inequality
2160 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 | 795 | 210 | 105 | 955 | 2131 |
| Governance & Regulation | 886 | 414 | 197 | 126 | 1654 |
| Organizational Efficiency | 826 | 204 | 129 | 87 | 1257 |
| Technology Adoption Rate | 681 | 259 | 128 | 110 | 1189 |
| Research Productivity | 464 | 138 | 65 | 349 | 1028 |
| Output Quality | 503 | 196 | 61 | 53 | 813 |
| Decision Quality | 351 | 180 | 84 | 51 | 673 |
| AI Safety & Ethics | 238 | 288 | 71 | 34 | 637 |
| Firm Productivity | 455 | 58 | 92 | 20 | 631 |
| Market Structure | 186 | 172 | 123 | 25 | 511 |
| Task Allocation | 222 | 70 | 76 | 34 | 407 |
| Innovation Output | 238 | 28 | 48 | 18 | 334 |
| Skill Acquisition | 177 | 62 | 62 | 17 | 318 |
| Employment Level | 107 | 57 | 108 | 13 | 287 |
| Fiscal & Macroeconomic | 135 | 72 | 44 | 26 | 284 |
| Firm Revenue | 172 | 50 | 28 | 5 | 256 |
| Consumer Welfare | 121 | 68 | 45 | 12 | 246 |
| Task Completion Time | 183 | 33 | 10 | 13 | 240 |
| Inequality Measures | 45 | 126 | 50 | 6 | 227 |
| Worker Satisfaction | 95 | 74 | 23 | 12 | 204 |
| Error Rate | 77 | 98 | 11 | 4 | 190 |
| Regulatory Compliance | 84 | 73 | 17 | 7 | 181 |
| Automation Exposure | 61 | 61 | 27 | 14 | 166 |
| Training Effectiveness | 98 | 21 | 14 | 19 | 154 |
| Wages & Compensation | 78 | 37 | 25 | 6 | 146 |
| Developer Productivity | 105 | 18 | 14 | 6 | 144 |
| Team Performance | 87 | 17 | 28 | 10 | 143 |
| Job Displacement | 12 | 83 | 23 | 1 | 119 |
| Hiring & Recruitment | 53 | 8 | 8 | 3 | 72 |
| Social Protection | 39 | 17 | 8 | 2 | 66 |
| Creative Output | 32 | 20 | 8 | 3 | 64 |
| Skill Obsolescence | 5 | 50 | 6 | 1 | 62 |
| Labor Share of Income | 17 | 20 | 17 | — | 54 |
| Worker Turnover | 15 | 15 | — | 3 | 33 |
| Industry | — | — | — | 1 | 1 |
Stress tests confirmed scalability: solver times remained under 95 seconds for instances with 1,000 staff members.
Scalability/stress testing reported in the paper using scheduling solver on problem instances with up to 1,000 staff; hardware and solver configuration not specified in the excerpt.
The performance evaluation framework analysis revealed 74% positive patient feedback.
Reported result from NLP analysis of patient surveys in the experiments; the number of patient survey responses and timeframe are not provided in the excerpt.
The intelligent staff scheduling module reduces scheduling conflicts by 41% compared to conventional methods while improving fairness (Gini coefficient = 0.08).
Results from scheduling optimization experiments reported in the paper; comparison against unspecified 'conventional methods'; specific experimental sample sizes (number of staff/rosters used for the comparison) not provided in the excerpt.
Workforce demand forecasting using LSTM, XGBoost, and Random Forest models predicts patient admissions and staffing needs, with LSTM achieving the best performance (MAE = 6.1, R2 = 0.91).
Experimental comparison of ML models on synthetic and real hospital datasets; reported forecasting metrics MAE and R2 for LSTM (other models' metrics not quoted in the provided text). The specific dataset size and train/test splits are not reported in the excerpt.
The study's findings provide strategic guidance for firms seeking long-term sustainable growth through reliance on generative AI to improve ESG performance.
Interpretation and managerial implications drawn from the empirical results of the panel analyses (2012–2024 Chinese A-share sample); presented as implications/recommendations in the paper's discussion section.
Hybrid professional competencies — combining digital and AI literacy, transversal (soft) skills, and ethical oversight capabilities — are necessary in AI-driven environments.
Consolidated finding from accreditation journal sources analyzed via thematic content analysis in the qualitative library research (number and identity of sources not specified).
Sustainable adaptation to AI requires continuous upskilling and reskilling ecosystems supported by organizations and policymakers.
Recommendation drawn from thematic synthesis of policy and organizational literature reviewed in the study (qualitative review; no quantified samples provided).
AI supports innovative work models such as human–AI collaboration.
Thematic synthesis of journal sources discussing AI adoption and work models in the qualitative library research (number of sources unspecified).
AI increases productivity.
Consolidated evidence from recent peer-reviewed studies included in the qualitative literature review (specific studies and sample sizes not listed).
AI generates new job categories.
Synthesis of findings from accredited journal articles reviewed in the library research (study design: literature analysis; sample size of articles not provided).
The positive impact of DDDM on international firm performance is amplified by state ownership.
Reported interaction/moderation result in the paper indicating that state ownership increases the strength of the DDDM–performance relationship (specific empirical details not provided in the excerpt).
The positive impact of DDDM on international firm performance is amplified by greater foreign shareholding.
Reported interaction/moderation finding in the paper showing that higher foreign shareholding enhances the positive DDDM–performance effect (detailed statistics and sample description not included in the excerpt).
The positive impact of DDDM on international firm performance is amplified by higher market competition.
Reported interaction/moderation result in the paper indicating that market competition strengthens the DDDM–performance relationship (specific interaction coefficients, significance levels, and sample details not provided in the excerpt).
DDDM positively relates to sustainability vision co-creation (future external).
Listed in the paper's framework as the future external dimension through which DDDM generates sustainable value and influences performance (empirical backing not specified in the excerpt).
DDDM positively relates to sustainability information disclosure (current external).
Identified as a current external mechanism in the paper's framework linking DDDM to improved international firm performance (supporting analyses not detailed in the excerpt).
DDDM positively relates to green innovation (future internal).
Included in the paper's framework as one of the four mechanisms through which DDDM creates sustainable value and affects firm performance (empirical support details not provided in the excerpt).
DDDM positively relates to pollution prevention (current internal) activities.
Part of the paper's framework and reported findings tying DDDM to the 'pollution prevention' dimension (empirical support details not included in the excerpt).
DDDM creates sustainable value for firms and thereby enhances international firm performance across four dimensions: pollution prevention (current internal), green innovation (future internal), sustainability information disclosure (current external), and sustainability vision co-creation (future external).
The paper presents a developed conceptual/framework explanation linking DDDM to sustainable value creation across the four specified dimensions; the excerpt does not specify whether these links are supported by mediation analysis or qualitative/theoretical argumentation.
Data-driven decision-making (DDDM) positively impacts international firm performance.
Empirical analysis reported in the paper in which DDDM is quantified using AI language models (BERT and ChatGLM2-6B) and related statistically to measures of international firm performance (details on sample size and statistical tests not provided in the excerpt).
AI-supported HR processes would have produced measurable increases in output per worker (labor productivity).
Counterfactual simulations and predictive estimates from the industrial firm dataset projecting output per worker under AI-HRM scenarios.
AI-HRM would have led to better alignment between training and production needs (improved targeting of training intensity to production requirements).
Model links training intensity to production outcomes and projects improved training–production alignment under AI-supported HR processes via regression-based simulations. (Quantitative magnitudes not specified in the description.)
Firms characterized by high labor intensity, rigid hierarchical structures, and limited coordination mechanisms would have experienced the strongest efficiency and productivity gains under an AI-HRM scenario.
Heterogeneity analysis within the regression-based simulation results from the industrial firm dataset (counterfactual projections by firm-type characteristics). (Details on how many firms fell into each category not provided.)
AI-driven HRM (AI-HRM) could have increased organizational efficiency and workforce performance (profitability, operational efficiency, defect reduction, and total output) in historical industrial firms.
Counterfactual analytical model built from an industrial firm dataset; regression-based simulations and predictive estimation linking HR indicators to organizational outcomes. (Dataset sample size and period not specified in the description.)
Findings reinforce behavioral economics perspectives on bounded rationality and adaptive performance.
Authors interpret results as aligning with behavioral economics concepts (bounded rationality, adaptive performance). This is an interpretive claim drawn from the study's empirical patterns; no direct tests of bounded rationality are described in the excerpt.
Ensemble machine learning models outperform traditional approaches in this behavioral and labor economics analysis.
Methodological claim in the paper: ensemble ML models were compared to traditional approaches and reported to outperform them. The excerpt does not provide performance metrics (e.g., R^2, RMSE, accuracy), cross-validation details, or sample size.
Productivity gains are realized through sustained mental health and active work involvement rather than isolated skill acquisition.
Interpretation based on mediation findings reported by the authors showing wellbeing and engagement channels; no quantitative comparisons or sample details are provided in the excerpt to quantify the contrast with isolated skill acquisition.
Psychological well-being and work engagement significantly mediate the relationships between emotional/psychological traits and productivity.
Study reports mediation analysis results where psychological well-being and work engagement serve as mediators in the machine-learning analysis. Details on mediation method, sample size, and significance statistics are not provided in the excerpt.
Emotional intelligence is a dominant predictor of labor productivity, outperforming personality traits, AI literacy and work environment factors.
Reported result from the study's analysis using a machine-learning based analytical approach (ensemble models). Variables included emotional intelligence, personality traits, AI literacy, and work environment factors. Specific sample size, effect sizes, and statistical metrics are not provided in the text excerpt.
Large language models (LLMs) perform reliably when their outputs can be checked (examples: solving equations, writing code, retrieving facts).
Statement in paper supported by illustrative examples (equations, code, factual retrieval); no large-scale quantitative benchmark reported in the abstract; evidence appears to be qualitative/anecdotal within the paper.
AI-assisted tools have shown promise in improving detection rates and workflow efficiency in gastroenterology.
Background statement referencing prior work and the studies included in the review that reported improved detection and efficiency.
AI reduced reading time by 30% in some studies.
Reported finding in the review summarizing time-efficiency outcomes from subset of included studies (magnitude reported as 30% reduction).
Among 40 included studies, AI demonstrated high diagnostic accuracy, with sensitivities and specificities exceeding 90% in lesion detection.
Aggregate result reported in the review summarizing diagnostic performance across the 40 included studies.
Findings provide granular evidence to support differentiated regional and industrial policies aimed at strengthening supply chain resilience.
Policy implication derived from heterogeneity analyses (ownership, industry, region) on the 2012–2022 Shanghai and Shenzhen A-share dataset.
The paper empirically clarifies the previously opaque ('black-box') mediation role of technological innovation between NQPF and supply chain efficiency.
Use of mediating-effect models on 2012–2022 A-share panel data to quantify mediation (including reported mediation proportion of 84.6%).
This study develops a unified NQPF theoretical framework integrating digital, green, and talent dimensions.
Authors' stated theoretical integration in the paper, presenting a multi-dimensional NQPF framework combining digital, green, and talent elements.
NQPF’s positive impact on supply chain efficiency is stronger in Eastern China compared with other regions.
Regional heterogeneity analysis using the 2012–2022 A-share panel data showing larger estimated effects for firms located in Eastern China.
The positive effect of NQPF on supply chain efficiency is stronger in state-owned enterprises (SOEs) than in non-state firms.
Heterogeneity analysis by ownership type performed on the 2012–2022 A-share panel data showing larger coefficients/effects for SOEs.
NQPF affects supply chain efficiency via multiple mechanisms: technological innovation, management restructuring, and digital transformation.
Mechanism analysis using mediating-effect models and supplementary tests on the 2012–2022 A-share panel data identifying these specific mediators.
Population growth shows a significant positive effect on GDP growth across the countries in the sample.
Population growth entered as a regressor and reported significant positive association with GDP growth in the panel models (OLS, FE, Difference and System GMM); exact magnitude and significance levels not provided in the summary.
Government expenditure shows a significant positive effect on GDP growth across the countries in the sample.
Positive and statistically significant coefficients on government expenditure reported in the applied econometric models (OLS, FE, Difference and System GMM); government spending included as a control macroeconomic determinant (sample/time not specified).
Gross fixed capital formation (GFCF) has a significant positive effect on GDP growth across the countries in the sample.
Estimated positive and statistically significant coefficients on GFCF in the panel regressions (OLS, FE, Difference and System GMM); GFCF included as a macroeconomic determinant in the model (sample size/time period not provided).
The combination of incentive-mediated adaptive interaction and persistent environmental memory can produce 'intelligent' coordination dynamics (structured, viable coordination behaviors) without assuming welfare maximization, rational expectations, or centralized design.
Synthesis claim supported by the above theoretical results (existence of bounded invariant sets, non-reducibility to global objectives, history sensitivity, and linear examples showing varied dynamical regimes). The evidence is theoretical/examples rather than empirical.
By mapping current evidence and identifying critical barriers, this review provides a foundational roadmap for researchers, policymakers, and practitioners aiming to leverage AI for inclusive economic growth in Jaipur’s micro‑enterprise sector.
Authors' concluding claim about the contribution of the review based on synthesized findings and identified barriers; presented as the paper's intended utility.
Targeted interventions—such as subsidized AI training programs, public–private partnerships to upgrade micro‑enterprise infrastructure, and gender‑responsive regulatory policies—are necessary to realize AI’s full benefits for women entrepreneurs.
Authors' recommendations derived from the review findings (identification of barriers leads to proposed interventions); recommendations presented as remedies to the synthesized gaps.
AI enables flexible, remote work arrangements that better accommodate women’s socio‑cultural needs.
Synthesis of qualitative and/or quantitative evidence in the included articles indicating AI‑enabled remote/flexible work arrangements and their fit with socio‑cultural constraints affecting women entrepreneurs.
AI tools significantly improve workflow productivity, for example reducing manual processing time by up to 40%.
Quantitative findings aggregated or cited within the included studies as synthesized in the review; the paper reports an example figure of 'up to 40%' reduction in manual processing time drawn from the literature.
The study presents a complementary linking theory that connects sustainability practice and reasoning to inform future discourse on sustainable e-commerce growth strategy in the dual carbon phase.
Theoretical/conceptual contribution described in the paper; this is a conceptual claim rather than an empirical finding.
Recognition of the gender dimensions of social protection has grown over recent decades, and program designs and research questions have evolved to explicitly address gender issues.
Descriptive claim about trends over time stated by the authors; implied support from evolving policy/program designs and research agendas, but no specific trend-data or studies cited in the excerpt.
Gender considerations in the design and delivery of programs are critical for social protection to achieve its primary objectives of reducing poverty and vulnerability.
Assertion in the chapter introduction; authors state this as a general principle to be examined. No specific empirical method or sample size cited in the excerpt (the chapter uses a 'review of reviews' approach to summarize evidence).
The study offers culturally sensitive, scalable strategies for policymakers, workforce agencies, and employers that improve immigrant integration, foster equitable labor market participation, and reduce structural inequalities.
Policy and practice recommendations derived from mixed-methods findings (survey n=150; interviews n=70 total) and comparative evaluation of translation models; recommendations reported in the paper's practical implications.