Evidence (7448 claims)
Adoption
5267 claims
Productivity
4560 claims
Governance
4137 claims
Human-AI Collaboration
3103 claims
Labor Markets
2506 claims
Innovation
2354 claims
Org Design
2340 claims
Skills & Training
1945 claims
Inequality
1322 claims
Evidence Matrix
Claim counts by outcome category and direction of finding.
| Outcome | Positive | Negative | Mixed | Null | Total |
|---|---|---|---|---|---|
| Other | 378 | 106 | 59 | 455 | 1007 |
| Governance & Regulation | 379 | 176 | 116 | 58 | 739 |
| Research Productivity | 240 | 96 | 34 | 294 | 668 |
| Organizational Efficiency | 370 | 82 | 63 | 35 | 553 |
| Technology Adoption Rate | 296 | 118 | 66 | 29 | 513 |
| Firm Productivity | 277 | 34 | 68 | 10 | 394 |
| AI Safety & Ethics | 117 | 177 | 44 | 24 | 364 |
| Output Quality | 244 | 61 | 23 | 26 | 354 |
| Market Structure | 107 | 123 | 85 | 14 | 334 |
| Decision Quality | 168 | 74 | 37 | 19 | 301 |
| Fiscal & Macroeconomic | 75 | 52 | 32 | 21 | 187 |
| Employment Level | 70 | 32 | 74 | 8 | 186 |
| Skill Acquisition | 89 | 32 | 39 | 9 | 169 |
| Firm Revenue | 96 | 34 | 22 | — | 152 |
| Innovation Output | 106 | 12 | 21 | 11 | 151 |
| Consumer Welfare | 70 | 30 | 37 | 7 | 144 |
| Regulatory Compliance | 52 | 61 | 13 | 3 | 129 |
| Inequality Measures | 24 | 68 | 31 | 4 | 127 |
| Task Allocation | 75 | 11 | 29 | 6 | 121 |
| Training Effectiveness | 55 | 12 | 12 | 16 | 96 |
| Error Rate | 42 | 48 | 6 | — | 96 |
| Worker Satisfaction | 45 | 32 | 11 | 6 | 94 |
| Task Completion Time | 78 | 5 | 4 | 2 | 89 |
| Wages & Compensation | 46 | 13 | 19 | 5 | 83 |
| Team Performance | 44 | 9 | 15 | 7 | 76 |
| Hiring & Recruitment | 39 | 4 | 6 | 3 | 52 |
| Automation Exposure | 18 | 17 | 9 | 5 | 50 |
| Job Displacement | 5 | 31 | 12 | — | 48 |
| Social Protection | 21 | 10 | 6 | 2 | 39 |
| Developer Productivity | 29 | 3 | 3 | 1 | 36 |
| Worker Turnover | 10 | 12 | — | 3 | 25 |
| Skill Obsolescence | 3 | 19 | 2 | — | 24 |
| Creative Output | 15 | 5 | 3 | 1 | 24 |
| Labor Share of Income | 10 | 4 | 9 | — | 23 |
Robustness checks include mediator tests (costs, tariffs, logistics) and firm‑level subgroup analyses to establish heterogeneous responses and support mechanism claims.
Paper reports robustness strategy involving mediation analysis and subgroup DID estimations across multiple mediator variables and firm size groups using the stated databases.
Empirical identification relies on treating CAFTA as an exogenous shock and applying a difference‑in‑differences (DID) design on firm and customs data from 2000–2014.
Methodological description in the paper: DID strategy with treated vs control comparisons; data sources explicitly listed as the China Industrial Enterprise Database and China Customs Database covering 2000–2014.
Highly Autonomous Cyber-Capable Agents (HACCAs) are AI systems able to plan and execute multi-stage cyber campaigns across the full attack lifecycle with minimal or no human direction.
Conceptual definition provided in the report; constructed via literature review and threat-framework formulation (no empirical sample; definitional/analytic).
Potential risks of deploying such models include fairness/bias, privacy concerns from employee-level predictions, and adverse morale effects if interventions are unevenly applied.
Authors' discussion of risks and ethical considerations when applying predictive XAI models to employee data; this is a stated limitation/risk discussion rather than an empirical finding.
Generalizability is limited: results based on the IBM dataset may differ for real green-workforce populations, industries, or countries.
Authors' stated limitation regarding external validity and representativeness of the IBM HR Analytics dataset as a proxy for sustainability roles.
Counterfactual simulations reported are predictive rather than causal; estimated effects require causal validation (e.g., randomized trials) before large-scale policy rollout.
Authors' methodological caveat noting that simulation-based changes in model-predicted probabilities do not establish causality and recommending causal evaluation methods for policy adoption.
The IBM HR Analytics dataset was used as a proxy for sustainability-focused (green) roles, relying on objective HR records rather than self-report surveys.
Data statement in the paper: model trained and evaluated on the IBM HR Analytics dataset; authors explicitly treat it as a proxy for sustainability-oriented roles for purposes of demonstration.
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.
Practical recommendations for firms and policymakers include investing in training for AI curation/evaluation/coordination, experimenting with decentralised decision rights and governance safeguards, and monitoring competitive dynamics related to model/platform providers.
Policy and practitioner takeaways explicitly presented in the discussion/implications sections, deriving from the conceptual framework and mapped literature.
The paper recommends a research agenda for AI economists: causal microeconometric studies (DiD, IVs, RCTs), structural models with hybrid human–AI agents, measurement work on GenAI use, distributional analysis and policy evaluation.
Explicit recommendations listed in the implications and research agenda sections; logical follow‑on from bibliometric findings about gaps in causal and measurement evidence.
Bibliometric mapping profiles the intellectual structure and evolution of the field but does not establish causal effects of GenAI on organisational outcomes.
Methodological limitation explicitly stated in the paper; bibliometric approach (co‑word, citation, thematic mapping) is descriptive and historical in scope.
Co‑word and thematic analyses reveal six coherent conceptual clusters that bridge technical AI topics (e.g., LLMs, GANs) with managerial themes (e.g., autonomy, coordination, decision‑making).
Thematic mapping and co‑word network analysis performed on the 212‑paper corpus; identification of six clusters reported in results.
Bibliometric and conceptual tools (VOSviewer, Bibliometrix) were used to identify performance trends, co‑word structures, thematic maps, and conceptual evolution in the GenAI–organisation literature.
Methods section: use of VOSviewer for network visualization and Bibliometrix for bibliometric statistics, co‑word analysis, thematic mapping and Sankey thematic evolution.
The study analysed a corpus of 212 Scopus‑indexed publications covering 2018–2025 to map emergent literature on Generative AI and organisational change.
Bibliometric dataset constructed from Scopus; sample size = 212 peer‑reviewed articles; time window 2018–2025; analyses performed with Bibliometrix and VOSviewer.
Research agenda: causal studies (panel data, quasi-experiments) are needed to estimate effects of AI exposure on employment outcomes and to evaluate retraining/income-support interventions for pre-retirement populations.
Authors’ stated recommendation based on limits of cross-sectional regression results from the n=889 survey and the identified need to move from association to causation.
Study limitations: cross-sectional design, self-reported intentions, potential unobserved confounders, and limited generalizability to only three cities (Beijing, Guangzhou, Lanzhou).
Explicit methodological statements in the paper describing data and design: cross-sectional survey of 889 respondents from three cities and reliance on self-reported employment intentions.
Because the study is cross-sectional and self-report, causal claims are limited and generalizability is restricted to Generation Z (limitation noted in the paper).
Authors' limitations: cross-sectional/self-report design and sample restricted to Generation Z; these constraints are reported in the paper.
Study design: cross-sectional self-report survey of 450 Generation Z consumers analyzed with Structural Equation Modeling (SPSS AMOS).
Methods section reporting sample size (n = 450), target population (Generation Z), cross-sectional survey design, and analysis technique (SEM using SPSS AMOS).
The measurement and structural model show good to excellent fit and reliable constructs (CFI = 0.980, TLI = 0.974, RMSEA = 0.062, SRMR = 0.031).
Reported psychometric/model-fit indices from SEM analysis (SPSS AMOS) on sample of 450 respondents.
Outcomes reported are primarily self-reported psychological measures rather than objective productivity metrics.
Paper reports measurement instruments focused on self-reported self-efficacy, psychological ownership, meaningfulness, and enjoyment/satisfaction; no primary objective productivity metrics reported.
The experiment was pre-registered, used occupation-specific writing tasks, and employed a between-subjects design with three conditions (No-AI, Passive AI, Active collaboration).
Study design reported in the paper: pre-registration statement, N = 269, between-subjects assignment to three conditions using occupation-specific writing tasks.
Active, collaborative AI use preserves perceived meaningfulness of work at levels comparable to independent work and does not produce the lasting psychological costs seen with passive use.
Pre-registered experiment (N = 269) with post-manipulation and post-return measures; Active-collaboration condition matched No-AI on meaningfulness and showed no persistent declines after returning to manual tasks.
Active, collaborative AI use preserves psychological ownership of outputs at levels comparable to independent work.
Pre-registered experiment (N = 269); Active-collaboration condition reported ownership levels similar to No-AI condition on self-report scales.
Active, collaborative AI use (human drafts first, then uses AI to refine) preserves self-efficacy at levels comparable to independent (no-AI) work.
Pre-registered experiment (N = 269) comparing Active-collaboration and No-AI conditions; no statistically meaningful differences in self-efficacy between them (self-reported measures).
The paper identifies future research directions, including empirical causal studies on how DPP+AI interventions change recycling rates, second‑hand market prices, and firm investment in circular processes; and modeling firm strategy around proprietary vs shared DPP data.
Stated research agenda and gaps in the paper informed by the study's findings and limitations; these are recommendations rather than empirical claims.
The study used a mixed-methods design focused on the Italian fashion and cosmetics industries, employing two online surveys, k‑means clustering (consumer segmentation), principal component analysis (to identify underlying dimensions of DPP functionalities and sustainability practices), and logistic regression (to identify adoption drivers).
Methods section summary provided in the paper; explicit statement of methods and industry context. Note: sample sizes and survey instrument details are not provided in the summary.
Two consumer segments were identified: 'aware' consumers (environmentally attuned and receptive to digital innovation and sustainability information) and 'unaware' consumers (prioritize immediate, tangible benefits like price and convenience over sustainability information).
K‑means cluster analysis applied to consumer responses from one of the online surveys in the Italian fashion and cosmetics context; summary identifies two clusters; sample sizes not reported.
This work is a conceptual/policy analysis rather than an original empirical study.
Explicit statement in the paper's Data & Methods section.
Study limitations include single-country (China) listed‑firm sample and reliance on secondary/administrative proxies for digitalization and innovation, which may miss internal qualitative aspects and introduce measurement error.
Authors’ stated limitations: sample restricted to Chinese A-share listed firms (2012–2022) and measures of digitalization/innovation derived from administrative/secondary data rather than direct observation/survey of internal practices.
No new primary empirical tests were performed in this paper; conclusions are based on secondary analysis and are broad and diagnostic rather than demonstrating causal mechanisms.
Explicit methodological statement in the Data & Methods and Limitations sections of the paper describing it as a qualitative literature review and synthesis.
Research should prioritize causal identification (IV, difference‑in‑differences, regression discontinuity) to disentangle whether ESG causes better financial outcomes or instead proxies for unobserved firm quality.
Methodological recommendation based on limitations in the reviewed literature (many observational/correlational studies); the paper argues for stronger causal designs going forward.
The authors propose research priorities for economists: quantify productivity gains from closing the actionability gap; estimate firm-level heterogeneity in evaluation capability and its effect on adoption; and model investment trade-offs between building evaluation-to-action pipelines versus accepting reduced LLM performance.
Paper's concluding recommendations for future research directions (explicitly listed by the authors).
The paper produces as primary outcomes a taxonomy of ten evaluation practices, the articulation of the results-actionability gap, and recommended strategies observed among successful teams.
Authors report these as the main outcomes of their thematic analysis and syntheses from the 19 interviews.
The study method consisted of semi-structured qualitative interviews with 19 practitioners across multiple industries and roles, analyzed via thematic coding.
Explicit methods section of the paper stating sample size (n=19), participant diversity, interview approach, and coding/analysis procedure.
AI-economics research should treat quantum capability as a distinct, gradually diffusing factor of production with sectoral specificity and model complementarities and policy counterfactuals endogenously.
Modeling recommendations grounded in sensitivity of macro outcomes to diffusion patterns, complementarities, and policy choices observed in the scenario and counterfactual analyses.
Model parameters are calibrated using historical diffusion of enabling technologies (cloud computing, GPUs, AI toolchains), industry case studies, and expert elicitation where hard data are lacking.
Empirical grounding section describing calibration sources: historical diffusion, case studies (materials discovery, optimization), and expert elicitation.
Uncertainty quantification is performed by running Monte Carlo or scenario ensembles and conducting sensitivity and robustness checks.
Methodological claim in the uncertainty quantification section describing Monte Carlo/scenario ensemble approach.
Sectoral TFP shocks are integrated into computational general equilibrium (CGE) or multi-sector growth models (and optionally DSGE variants) to simulate GDP, sector output, trade impacts, and labor reallocation.
Method section stating integration of sectoral TFP shocks into CGE/multi-sector growth models with optional DSGE short-run dynamics.
Sectoral adoption is translated into total factor productivity (TFP) shocks or sector-specific Hicks-neutral productivity improvements based on micro evidence of quantum advantages.
Methodological description of productivity mapping linking adoption to TFP shocks using micro evidence and case studies.
The paper uses empirical diffusion functions (logistic/S-curve, Bass model) calibrated to analogous technologies to project uptake over time.
Methodological description: diffusion modeling section explicitly states use of logistic/S-curve and Bass models and calibration to past technologies (cloud, GPUs).
The analysis used sentence‑transformer models to produce dense vector representations of article text and UMAP to project those embeddings into a low‑dimensional thematic map for cluster identification and gap detection.
Methods section specifying use of sentence‑transformer embeddings and UMAP for dimensionality reduction/visualization of article text.
The study followed a PRISMA protocol for literature selection and included peer‑reviewed journal articles published between 2014 and 2024, with a final sample size of n = 109.
Explicit methodological statement in the paper describing the literature search, inclusion/exclusion criteria, and final sample.
Twenty‑seven papers study marketing in banking without using NLP methods.
PRISMA systematic review; categorization of the 109 selected articles into the three coverage groups (8, 74, 27).
Seventy‑four papers study NLP in marketing more broadly (not specifically banking).
Same PRISMA‑based systematic review and manual categorization of the final sample n = 109 into topical buckets (NLP in marketing vs. NLP in bank marketing vs. marketing in banking without NLP).
Only 8 peer‑reviewed papers directly examine NLP in bank marketing (out of a final sample of 109 articles published 2014–2024).
Systematic review following PRISMA protocol; final sample n = 109 peer‑reviewed journal articles published 2014–2024; manual screening and categorization yielding counts by topic.
The study's findings are qualitative and case-driven (Xiaomi and Deloitte); generalizability is limited by case selection and the absence of standardized quantitative metrics.
Methods section explicitly states case analysis and literature review as primary methods and notes lack of large-scale quantitative measurement.
The methodology is normative-philosophical argumentation supplemented by interdisciplinary synthesis (phenomenology, deconstruction, OOO, STS/material turn); this is not an empirical causal study and contains no quantitative datasets.
Author-declared methods and limits: statement that the intervention is theory-driven and qualitative; absence of quantitative analysis reported.
The paper’s empirical grounding consists of illustrative case studies and vignettes from healthcare robotics, autonomous vehicles, and algorithmic governance used to demonstrate distributed agency and responsibility.
Author-stated methodology: qualitative vignettes/case illustrations across three domains; no reported sample sizes or systematic data collection.
The analysis in the paper is primarily qualitative and descriptive; it does not empirically quantify AI’s effects on trade flows or welfare.
Explicit statement in the methods/data description noting a mixed qualitative approach (theoretical analysis, comparative legal analysis, case studies, scenario reasoning) and absence of empirical quantification.