Evidence (254 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
10085 claims
Filter claims →
Productivity
8974 claims
Filter claims →
Governance
8062 claims
Filter claims →
Human-AI Collaboration
7749 claims
Filter claims →
Org Design
5057 claims
Filter claims →
Innovation
4896 claims
Filter claims →
Labor Markets
4088 claims
Filter claims →
Skills & Training
3372 claims
Filter claims →
Inequality
2377 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 | 882 | 244 | 117 | 1097 | 2424 |
| Governance & Regulation | 1010 | 469 | 229 | 135 | 1875 |
| Organizational Efficiency | 977 | 235 | 149 | 90 | 1462 |
| Technology Adoption Rate | 781 | 299 | 143 | 128 | 1362 |
| Research Productivity | 506 | 155 | 74 | 363 | 1110 |
| Output Quality | 555 | 219 | 71 | 70 | 915 |
| Decision Quality | 395 | 200 | 95 | 54 | 751 |
| Firm Productivity | 523 | 67 | 101 | 27 | 724 |
| AI Safety & Ethics | 262 | 309 | 75 | 36 | 688 |
| Market Structure | 195 | 201 | 135 | 30 | 566 |
| Task Allocation | 248 | 77 | 96 | 38 | 464 |
| Innovation Output | 300 | 34 | 55 | 20 | 411 |
| Skill Acquisition | 207 | 75 | 65 | 21 | 368 |
| Employment Level | 138 | 67 | 119 | 24 | 350 |
| Fiscal & Macroeconomic | 156 | 80 | 53 | 33 | 329 |
| Task Completion Time | 211 | 38 | 13 | 16 | 280 |
| Firm Revenue | 183 | 52 | 29 | 5 | 270 |
| Consumer Welfare | 131 | 77 | 48 | 13 | 269 |
| Inequality Measures | 50 | 141 | 54 | 9 | 254 |
| Worker Satisfaction | 104 | 85 | 25 | 13 | 227 |
| Error Rate | 87 | 112 | 11 | 5 | 215 |
| Automation Exposure | 69 | 69 | 37 | 20 | 198 |
| Wages & Compensation | 102 | 49 | 31 | 11 | 193 |
| Team Performance | 115 | 30 | 30 | 11 | 187 |
| Regulatory Compliance | 88 | 74 | 17 | 7 | 186 |
| Training Effectiveness | 109 | 22 | 14 | 21 | 168 |
| Developer Productivity | 116 | 21 | 15 | 8 | 161 |
| Job Displacement | 12 | 92 | 26 | 1 | 131 |
| Hiring & Recruitment | 57 | 12 | 9 | 5 | 83 |
| Skill Obsolescence | 6 | 59 | 10 | 2 | 77 |
| Social Protection | 43 | 17 | 8 | 2 | 70 |
| Creative Output | 35 | 21 | 9 | 4 | 70 |
| Labor Share of Income | 18 | 23 | 17 | 1 | 59 |
| Worker Turnover | 15 | 16 | — | 4 | 35 |
| Industry | — | — | — | 1 | 1 |
There are both similarities and differences in AI’s influence patterns on income distribution at different income levels (i.e., effects vary across overall inequality, top 1% share, and bottom 50% share).
Comparative fsQCA findings across the three outcome variables (Gini, top 1%, bottom 50%) reported by the authors for the N=20, T=5 panel.
AI produces different effects on income inequality depending on its compatibility with socioeconomic conditions (i.e., AI's effect is contingent on interacting socioeconomic factors).
Configurational fsQCA results showing multiple distinct AI-centered pathways and differing configurations of AI-related and socioeconomic conditions associated with different inequality outcomes in the N=20 sample.
The fsQCA analysis identifies 14 conditional configurations that are classified into seven types of AI-centered pathways affecting changes in income distribution.
Result reported from the sufficiency/configurational analysis of the fsQCA on the N=20, T=5 panel (paper reports 14 configurations and grouping into seven pathway types).
The effect of AI adoption on inequality is heavily moderated by a country's educational infrastructure and baseline economic development.
Reported moderation analysis / subgroup comparisons using OLS regression and Random Forest on the World Bank/OECD cross-country dataset indicating that the AI–inequality relationship varies with measures of education and development.
The comparative evaluation shows differences in economic inclusiveness between ML, DL, and Generative AI.
Abstract states differences in economic inclusiveness found in the review; no quantitative inclusiveness metrics or sample sizes provided in abstract.
AI functions as a conditional capability amplifier, expanding agency while producing uneven inclusion shaped by disparities in connectivity, skills, and infrastructure.
Analytical synthesis and illustrative empirical evidence from interviews showing differential effects tied to connectivity, skills, and infrastructure.
Overall, the digital economy brings both opportunities (raising incomes overall) and challenges (contributing to greater inequality).
Synthesis of empirical findings from the two-way fixed effects panel (31 provinces, 2011–2021) and robustness checks indicating positive average income effects alongside heterogeneous effects that widen disparities.
Regionally, eastern provinces experience greater income gains from digital development than central and western provinces.
Regional heterogeneity results from the paper's two-way fixed effects panel (31 provinces, 2011–2021) comparing estimated effects across eastern, central, and western regions.
Adverse employment and compensation effects are concentrated among workers in non-AI tasks and non senior-level positions, indicating an asymmetric distribution of gains from AI adoption.
Heterogeneity analysis / subgroup results showing larger negative employment/compensation responses for workers in non-AI tasks and for non senior-level positions across the sample.
Reward-level intervention (via equity-aware LLM refinement) significantly improves equity, but demographic disparities in AI-driven controllers persist.
Overall conclusion drawn from reported experimental results (improvements in group satisfaction metrics but acknowledgment that disparities remain).
Digital transformation has expanded connectivity and participation, but the benefits remain unevenly distributed due to asymmetries in data ownership, algorithmic governance, platform control, and value capture.
Argument supported by a literature review / conceptual synthesis of recent studies on digital transformation, data ownership, platform governance and value capture (no original empirical sample reported).
The distribution of gains from GenAI access was highly uneven across users.
Experimental results showing heterogeneous effects across participants (variance/heterogeneity analyses reported in the paper).
Adaptation determines who benefits from technological (AI) change.
One of five lessons; argued using historical analogy and labour market patterns (qualitative claim in chapter).
Aggregate effects are geographically uneven (geographic unevenness in AI-driven labor market impacts).
Synthesis across studies observing variation by geography and noting non-Anglophone markets and developing economies as under-studied and differentially affected.
Wage polarization characterizes the aggregate pattern of labor market change associated with recent AI advances.
Aggregate characterization from synthesized studies reporting divergent wage outcomes (higher wages for AI-augmented workers, pressures on junior/routine roles) consistent with polarization.
Introducing taxes on AI returns (τ_ai) and financial gains (τ_f) yields three distinct long-run regimes: low-tax (extreme inequality), moderate-tax (stable mixed economy), and high-tax (post-scarcity with universal basic income).
Model extension with tax parameters τ_ai and τ_f and analysis of steady states/long-run regimes; bifurcation analysis identifying regime types associated with ranges of (τ_ai, τ_f).
Automation reallocates income and ownership claims.
Theoretical model with heterogeneous households who hold capital/equity claims; equilibrium determines wages and returns and shows changes in income and ownership shares when automation increases.
While Agentic AI enhances economic performance, its benefits are mediated by structural conditions and are unevenly distributed across countries (i.e., reinforcing core–periphery inequalities).
Combined findings from fixed-effects regressions, mediation analysis, and observed heterogeneity between developed and emerging economies in the 2015–2024 panel.
Demographic characteristics intersect with AI exposure—i.e., exposure varies by demographic groups.
Paper reports that it examines how demographic characteristics intersect with exposure based on recent empirical studies; no demographic breakdowns or sample sizes provided in the abstract.
The distribution of complementary (non-AI) skills across the workforce shapes whether AI improvements generate productivity bottlenecks or concentration-driven inequality.
Derived from the task-based model analysis described in the article; framed as a theoretical mechanism with reference to empirical patterns but without specific empirical study details in the excerpt.
Model behaviors vary strongly with levels of reasoning and with users' inferred socio-economic status.
Reported findings from evaluations that varied model reasoning prompts/levels and user socio-economic status signals; paper states behavior differences across these dimensions. Abstract does not give sample sizes or exact quantitative differences.
The inequality-reducing impact of AI is weaker when carbon inequality is measured by the Theil index, implying persistent structural divides between advanced and less developed regions.
Same provincial panel dataset (2003–2021) with the Theil index as the dependent variable; results show a weaker (and impliedly less robust) association between AI development and Theil-measured carbon inequality.
Socioeconomic regression analysis confirms strong correlations between neighborhood racial composition and detection likelihood: Pearson r = 0.83 for percent White and r = -0.81 for percent Black.
Reported Pearson correlation coefficients from regression analysis between neighborhood racial composition variables and detection likelihood in the simulations.
A Conditional Tabular GAN (CTGAN) debiasing approach partially redistributes detection rates but cannot eliminate structural disparity without accompanying policy intervention.
Experimental comparison between baseline simulations and CTGAN-debiased synthetic data showing partial redistribution of detection rates; paper asserts remaining structural disparities.
Net gains from AI are not automatic nor evenly distributed; benefits depend on translation rates to clinical success and on addressing non-technical enablers.
Synthesis and conditional argument informed by sector observations; not backed by empirical distributional analysis in the paper.
There is substantial heterogeneity in worker experiences within platform-mediated gig work.
Observed variation in roles (primary vs. supplementary), earnings distribution (median below traditional but top-decile premiums), and access to benefits across the 24-country dataset from surveys, administrative records, and platform transaction data.
Whether AI increases or decreases overall inequality depends on AI’s technology structure (proprietary vs. commodity) and on labor-market institutions (rent‑sharing elasticity ξ and asset concentration).
Comparative statics and regime analysis within the calibrated model that varies the technological-form parameter (η1 vs. η0) and the rent‑sharing elasticity ξ, as well as measures of asset concentration.
AI can equalize individual task performance while increasing aggregate inequality because rents accrue to owners of complementary assets rather than to workers.
Analytical model and calibrated simulations demonstrating that within-task compression (reduced worker dispersion) can coexist with rising aggregate inequality (ΔGini) owing to rent concentration at the firm/asset-owner level.
Data‑driven policies can either amplify or mitigate inequalities depending on data representativeness, model design, and deployment governance.
Multiple empirical examples and theoretical analyses in the review highlighting cases of both harm (bias amplification) and mitigation, identified across the 103 items.
The benefits of digital expansion are concentrated among urban, better‑off and dominant‑group learners.
Comparative analysis using national datasets (UDISE+, AISHE, NSS, NFHS‑5, TRAI etc.) showing concentration of digital capital and access indicators in urban and higher socioeconomic/dominant caste groups.
Consequently, social impacts centre on heightened income inequality, deeper rural–urban and regional disparities, and persistent skill gaps, with transition costs falling on displaced garment, BPO, and customer-service workers.
Synthesis-derived conclusion from the paper combining sectoral exposure and demographic vulnerability; presented qualitatively without quantified measures.
Distributional heterogeneity is pronounced: women concentrated in export manufacturing and routine services are more exposed in many countries, though India’s agricultural employment profile produces an exception.
Paper reports gendered exposure patterns based on sectoral employment distributions from cited syntheses (ILO/World Bank); no numerical gender-disaggregated exposure rates provided.
Previous technological waves in the United States were largely routine- and skill-biased, contributing to job polarization and the concentration of labor-market power among highly educated, well-paid workers in nonroutine cognitive occupations.
Statement drawing on prior literature and historical analysis of earlier technology waves (routine- and skill-biased); no new empirical sample reported in the abstract.
The CAD has implications for knowledge-work stratification and AI platform governance.
Argumentative/policy discussion in the paper linking the CAD to potential stratification among knowledge workers and governance considerations for AI platforms.
AI can contribute to widening inequality.
Abstract reports the review identifies widening inequality as a potential trade-off of AI, based on synthesis of 194 articles.
AI adoption widens intra-firm pay disparities (increases pay inequality within firms).
Regression analyses showing divergent effects on employee vs. executive pay and explicit measures of intra-firm pay disparity in the panel data.
Developed economies leverage educational capital to mitigate the adverse inequality effects of AI adoption.
Reported interaction/moderation findings from OLS and Random Forest analyses on the World Bank/OECD dataset showing weaker or offset association between AI adoption and Gini in higher-education / higher-development country groups.
There is a global disparity in data centre infrastructure (concentrations favouring some regions over others).
Analysis drawing on external data sources cited in the paper illustrating geographic distribution of data centre infrastructure.
Policy asymmetries, digital literacy gaps, and regional inequalities deepen digital divides and impede inclusive development.
Policy analysis and comparative case studies documenting how policy differences, literacy, and regional disparities affect digital inclusion; China used as a focal example. No quantitative sample sizes or causal estimates given in summary.
The essay introduces the concept of a 'vouching gap' to describe a growing divide between students who graduate with credible advocates willing to stake their reputations on their behalf and those who do not.
Conceptual contribution defined in the essay and motivated by social capital theory and mentoring research; no empirical quantification or sample provided.
Automation of student work and candidate screening will widen existing inequalities between students.
Theoretical claim in the essay linking AI-driven automation to differential outcomes across students, motivated by social capital and mentoring literature; no empirical data or sample reported.
This automation threatens to hollow out the value of a university degree.
Argument presented in the essay, grounded in social capital theory and mentoring research; no empirical test or sample size reported.
Algorithmic systems produce inequitable outcomes for gig workers.
Interview data (16 workers, 21 stakeholders) reporting examples and perceptions of unequal treatment and distributional harms arising from algorithmic rules.
The analysis identifies four interrelated dynamics—algorithmic colonialism, data colonialism, platform imperialism, and platform sub-imperialism—through which dependency and domination are reproduced across global and intra-South contexts.
Synthesis of the 50 reviewed peer-reviewed articles; these dynamics are reported as the paper's analytical findings.
AI adoption may reproduce entrenched inequalities in postcolonial contexts.
Critical synthesis (literature review) of 50 peer-reviewed articles from 2019–2025 reported by the paper.
Unequal access to GenAI tools in higher education may exacerbate employability gaps and inequities among students.
Concern identified and discussed in the literature as summarized in the review article (conceptual/literature-based; no new empirical evidence reported).
Without intentional, gender‑aware interventions in policy and design, the AI‑driven gig economy is more likely to entrench existing social and economic inequalities than to alleviate them.
Conclusion and social implications in the paper based on thematic synthesis across 48 studies and the feminist political economy analysis.
AI‑enabled platforms reproduce and risk amplifying gender inequality through algorithmic bias, wage gaps, structural precarity, and digital marginalization.
Synthesis across the 48 reviewed studies identifying recurring mechanisms (algorithmic bias, wage gaps, precarity, digital marginalization) that disadvantage women; presented in Findings.
STARA may widen inequalities across occupational groups and cohorts—particularly affecting low- and medium-skill occupations—by fragmenting or limiting career paths and reducing institutional supports.
Concerns and literature synthesis in the editorial citing prior work on inequalities and occupational differences (e.g. Zajko, 2022 and other cited studies).
Algorithmic gatekeeping in promotion and evaluation processes can privilege certain behaviours or skill sets while limiting transparency and equity in career advancement.
Editorial synthesis referencing recent work (e.g. Hillebrand et al., 2025) and conceptual concerns raised in the literature.