Evidence (3308 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
9875 claims
Filter claims →
Productivity
8807 claims
Filter claims →
Governance
7870 claims
Filter claims →
Human-AI Collaboration
7560 claims
Filter claims →
Org Design
4892 claims
Filter claims →
Innovation
4781 claims
Filter claims →
Labor Markets
4004 claims
Filter claims →
Skills & Training
3308 claims
Filtered →
Inequality
2332 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 | 870 | 233 | 116 | 1066 | 2363 |
| Governance & Regulation | 976 | 451 | 218 | 133 | 1809 |
| Organizational Efficiency | 949 | 224 | 144 | 88 | 1416 |
| Technology Adoption Rate | 764 | 287 | 141 | 122 | 1325 |
| Research Productivity | 501 | 152 | 74 | 362 | 1101 |
| Output Quality | 542 | 216 | 69 | 69 | 896 |
| Decision Quality | 387 | 198 | 94 | 54 | 740 |
| Firm Productivity | 513 | 67 | 101 | 27 | 714 |
| AI Safety & Ethics | 249 | 303 | 73 | 36 | 667 |
| Market Structure | 190 | 192 | 134 | 27 | 548 |
| Task Allocation | 243 | 77 | 91 | 36 | 452 |
| Innovation Output | 291 | 33 | 55 | 20 | 401 |
| Skill Acquisition | 206 | 72 | 65 | 21 | 364 |
| Employment Level | 133 | 63 | 115 | 22 | 335 |
| Fiscal & Macroeconomic | 153 | 79 | 52 | 32 | 323 |
| Task Completion Time | 206 | 37 | 12 | 15 | 272 |
| Firm Revenue | 179 | 52 | 29 | 5 | 266 |
| Consumer Welfare | 130 | 76 | 47 | 13 | 266 |
| Inequality Measures | 48 | 137 | 51 | 6 | 242 |
| Worker Satisfaction | 101 | 81 | 25 | 13 | 220 |
| Error Rate | 84 | 110 | 11 | 5 | 210 |
| Wages & Compensation | 98 | 47 | 30 | 10 | 185 |
| Regulatory Compliance | 88 | 73 | 17 | 7 | 185 |
| Automation Exposure | 66 | 64 | 33 | 16 | 182 |
| Team Performance | 105 | 29 | 30 | 11 | 176 |
| Training Effectiveness | 109 | 22 | 14 | 21 | 168 |
| Developer Productivity | 114 | 21 | 14 | 8 | 158 |
| Job Displacement | 12 | 90 | 24 | 1 | 127 |
| Hiring & Recruitment | 57 | 9 | 9 | 5 | 80 |
| Skill Obsolescence | 6 | 56 | 9 | 1 | 72 |
| Social Protection | 43 | 17 | 8 | 2 | 70 |
| Creative Output | 35 | 21 | 9 | 4 | 70 |
| Labor Share of Income | 18 | 21 | 17 | 1 | 57 |
| Worker Turnover | 15 | 16 | — | 4 | 35 |
| Industry | — | — | — | 1 | 1 |
Skills Training
Remove filter
Mechanism analysis provides suggestive evidence that AI improves skill utilization and promotion expectations.
Mechanism/auxiliary analyses in the paper (described as 'suggestive evidence') linking AI diffusion to proxies for skill utilization and promotion expectations within the CLDS framework.
Effects of AI on the wage consequences of educational mismatch vary by occupation: AI mainly benefits overeducated workers in non-manual jobs, where surplus schooling can be effectively absorbed.
Subsample/heterogeneity analysis by occupation (manual vs. non-manual) in CLDS-linked fixed-effects models showing stronger attenuation of overeducation penalty in non-manual occupations under higher AI diffusion.
AI diffusion reduces the wage penalty for overeducated workers.
Interaction models using CLDS data and city-level AI diffusion in fixed-effects specifications showing that greater AI diffusion attenuates the negative wage effect for overeducated workers.
Undereducation is associated with a wage premium.
Same CLDS 2014–2018 microdata and cohort-based educational mismatch measure; estimated via extensive fixed-effects models showing higher wages for undereducated workers.
AI-related Policy and Government activity has a positive two-year lagged association with participation in education and training for adults aged 18-74 and 45-54.
Two-year lag specification in the structural panel models (two-way fixed effects) on 18 European countries, 2017–2024. Reported two-year lag coefficients: 0.6064 (ages 18-74) and 0.7346 (ages 45-54).
The most valuable asset a university can offer students in a post-AI economy is credible endorsement—the capacity of a trusted faculty member, advisor, or other mentor to vouch with specificity for a student's character, competence, and potential.
Normative/analytical claim in the essay based on social capital and mentoring research; presented as the author's recommended institutional response rather than empirically validated evidence.
Demand reallocates toward lower-priced workers in more AI-exposed job categories.
Empirical evidence from Upwork showing post-ChatGPT changes in hiring/demand patterns, with more AI-exposed categories shifting demand toward workers who charge lower prices; inferred via predictive models and difference-in-differences comparisons. (Sample size not reported in abstract.)
In more AI-exposed job categories, the importance of price in predicting labor demand rises.
Same empirical approach as above: Upwork data, text embeddings for worker profiles, computation of the predictive importance of price, and a difference-in-differences design around ChatGPT release comparing more- vs. less-AI-exposed categories. (Sample size not reported in abstract.)
Returnees are more likely than comparable stayers to receive unemployment benefits, and among recipients they receive higher daily benefit levels.
Chapter 4: linked Belgian administrative registers with benefit receipt and benefit-level information; causal comparisons between returnees and comparable stayers.
Augmentation AI stimulates program openings (new Bachelor programs) in exposed fields.
Chapter 3: supply-side analysis of program openings using U.S. higher-education program data 2010–2022; IV using lagged CS research intensity.
Augmentation AI attracts more and higher-ability students into exposed Bachelor programs.
Chapter 3: student demand and student-ability margins analyzed using U.S. enrollment/graduation data 2010–2022 and IV identification (lagged CS research intensity).
Augmentation AI increases Bachelor-degree graduations in AI-exposed fields in the U.S. (2010–2022).
Chapter 3: analysis of U.S. Bachelor program graduations 2010–2022; IV strategy using lagged computer-science research intensity to instrument AI exposure; margins include aggregate graduations and program supply.
AI-exposed occupations expand (grow) in employment/demand in the European sample.
Chapter 2: job-posting counts and occupational-level analyses from 75 million postings (2018–2023) across four countries, IV using lagged CS research intensity.
AI-exposed (AI, Data, Prediction) skills pair more frequently with complementary skill bundles (Judgment, Decision-Making, Leadership), i.e., increased co-occurrence of AI-exposed and complementary skills.
Chapter 2: co-occurrence analysis of extracted skills from 75 million job postings across four European countries (2018–2023), identified using multilingual skill classifiers and IV strategy.
In four European countries (2018–2023), there is significant growth in demand for AI, Data, and Prediction skills within AI-exposed occupations.
Chapter 2: analysis of 75 million online job postings across four European countries (2018–2023); multilingual skill extraction and classification using data-science methods; IV strategy instrumenting AI exposure with lagged computer-science research intensity.
Augmentation AI generates new work primarily for high-skilled occupations.
Chapter 1: occupational-level analysis using novel exposure measures and IV strategy (lagged CS research intensity) on U.S. data 2015–2022, heterogeneous by skill.
Augmentation AI raises wages primarily for high-skilled occupations.
Chapter 1: heterogeneous IV estimates by skill group using occupational exposure measures (Stack Overflow mapping and O*NET) for 2015–2022 U.S. data.
Automation AI increases employment in the U.S. (2015–2022).
Chapter 1: occupational exposure measures from Stack Overflow and O*NET; instrumental-variables strategy with lagged computer-science research intensity; U.S. labor-market data 2015–2022.
Augmentation AI stimulates the creation of new work in the U.S. (2015–2022).
Chapter 1: novel longitudinal measures of occupational exposure to augmentation AI constructed by mapping developer activity on Stack Overflow to occupational descriptions and an emerging-work measure from O*NET; instrumental-variables strategy using lagged computer-science research intensity; U.S. labor-market data covering 2015–2022.
The paper articulates testable propositions and discusses implications for organizational governance, AI system design, and institutional conditions that enable bounded openness and calibrated reliance.
Stated contributions in the abstract indicating policy/organizational/design implications and proposed propositions; descriptive claim about the paper's outputs rather than an empirical result.
LLM performance varies substantially with depth of disclosure, contextual richness, and iterative refinement — i.e., LLM output quality depends heavily on how users engage with them.
Theoretical claim and synthesis of related literatures (trust in automation, privacy calculus, algorithm aversion); no empirical method or sample sizes reported in the abstract.
Expert operators maintained a verification loop by persistently scanning the environment even when using LLM guidance.
Eye-tracking and behavioral data from expert participants showing continued environmental scanning (fixation metrics) and cross-referencing behavior in LLM-guided conditions.
LLM guidance enhanced task efficiency (higher rewards and victims-per-step) relative to a no-LLM baseline.
Experimental comparison in a simulated search-and-rescue environment across two LLM-guided conditions and a no-LLM baseline; behavioral measures reported for rewards and victims-per-step (eye-tracking and planning behavior also collected).
The paper derives C‑first policy prescriptions and offers three empirically testable propositions along with a falsifiable 10-year forecast.
Policy recommendations and empirical propositions presented in the paper (theoretical/policy-design evidence; forecast statement).
Convergence capacity (C) is distinct from absorptive capacity, dynamic capability, and human capital, and constitutes the specific cognitive mediator prior frameworks have left implicit.
Conceptual/definitional analysis and differentiation provided in the paper; theoretical argument distinguishing constructs.
A descriptive cross-national analysis of 20 OECD economies shows the AI × C interaction is associated with 86% of TFP variance, versus 31% for AI alone.
Empirical descriptive cross-national analysis reported in the paper; sample explicitly stated as 20 OECD economies (small-n analysis).
Using H-hat in the production function Y = F(K, H-hat) provides a human-centered mechanism for Solow's TFP residual: A_Solow = [1 + phi(A,C)]^(1-alpha).
Algebraic derivation connecting the proposed ICH augmentation factor to the Solow TFP residual (presented as a theoretical result in the paper).
The paper proposes the Intellectually Converged Human (ICH) framework with H-hat = H[1 + phi(A,C)], where effective productive capacity equals human capital (H) scaled by augmentation factor [1 + phi], and phi is jointly determined by AI utilization intensity (A) and convergence capacity (C).
Formal theoretical/model proposal presented in the paper (algebraic expression defining H-hat).
Adaptive measures from workers, employers, and governments are crucial to enable the new labor force to thrive under the future of AI.
Concluding synthesis/recommendation from the SLR that emphasizes multi-actor adaptation; no empirical effect estimates or intervention studies described in the excerpt.
Training centers, training programs, the education system, and governments have to adapt to help close the skills gap.
Policy/recommendation claim in the paper based on synthesis of reviewed studies that emphasize the need for institutional adaptation; specific policy evaluations or measured impacts not provided in the excerpt.
The review identifies new and emerging skills needed for jobs, such as data skills, machine learning skills, and digital communication skills, which the next generation of the workforce should have.
Synthesis from the SLR indicating recurring skills highlighted across empirical studies (specific studies, measures, and counts not provided).
By emphasizing mechanisms rather than direct effects, the dissertation offers a theoretically grounded and empirically supported explanation of how organizations convert digital technologies into sustained digital innovation.
Methodological/conceptual claim based on the dissertation's multi-study, mechanism-focused approach (Studies 1–3).
Digital innovation types (digital processes, digital products, digital business models) are distinct yet interconnected, and organizations generate them through capability chains in which digital culture, AI assimilation, and human-AI collaboration are pivotal.
Synthesis across Studies 1–3: theoretical argument supported by the empirical findings of Studies 2 and 3 (no quantitative summary or meta-analytic aggregation reported in the excerpt).
The translation of AI-related competence into digital business model innovation is shaped by boundary conditions, particularly digital leadership and AI-related financial resources (these variables moderate the competence -> AI assimilation -> business model innovation pathway).
Study 3: empirical findings reported as highlighting moderation by digital leadership and AI-related financial resources (exact moderation tests and sample size not provided in the excerpt).
Employee AI-related competence is translated into digital business model innovation only when an organizational mechanism — AI assimilation — converts dispersed competence into scalable and coordinated use (AI assimilation mediates the competence -> business model innovation relationship).
Study 3: empirical tests reported as supporting a mediating role for AI assimilation in the translation of employee AI competence into digital business model innovation (specifics of measurement, sample, and statistical tests not provided in the excerpt).
The relationship between AI and digital product innovation is largely indirect, operating through digital process innovation (i.e., AI -> process innovation -> product innovation).
Study 2: empirical findings described as demonstrating sequential mediation where AI's effect on digital product innovation occurs mainly via digital process innovation (specific method, mediation tests, and sample size not provided in the excerpt).
AI assimilation primarily drives digital process innovation, as AI reshapes workflows, decision processes, and coordination routines.
Study 2: empirical results reported as supporting AI assimilation as a driver of digital process innovation (specific empirical approach and sample size not provided in the excerpt).
Digital culture (openness to experimentation, collaboration, learning, and data-driven decision-making) enables AI to move beyond isolated experimentation and become embedded in organizational routines through AI assimilation.
Study 2: empirical test of a model linking digital culture to AI assimilation (described as empirical results supporting this link; specific method and sample size not provided in the excerpt).
I develop a comprehensive synthesis of digital innovation research in management and business and propose an overarching multilevel framework based on a Contextual Conditions - Mechanisms - Outcomes logic.
Study 1: literature review / intellectual-structure mapping of prior digital innovation research (comprehensive synthesis described; exact scope, databases, and sample of reviewed papers not reported in the excerpt).
The paper serves as a resource for policymakers and researchers addressing the economic and social impacts of robotics, artificial intelligence, and automation.
Stated in the paper's implications; reflects intended audience and utility rather than an empirical finding.
The study contributes to a limited body of research on robot taxation and offers guidance on adapting tax systems to technological change.
Claim about the paper's original contribution and scope, stated in the implications/originality/value section; based on the authors' review and synthesis of existing literature.
Implementing a robot tax approach supports responsible automation, reduces inequality, and fosters sustainable economic growth.
Conclusion/implication in paper based on synthesis of reviewed literature and normative argument; not presented as an empirically tested result within the study.
A robot tax would address tax policy biases that favour capital over labour.
Paper argues this normative point based on literature synthesis; presented as a rationale for the tax rather than proven empirically within the paper.
A robot tax could fund workforce retraining.
Policy recommendation in the paper deriving from the scoping review; framed as intended use of tax proceeds (no empirical trial or evaluation reported).
A robot tax is proposed to offset lost income tax revenue.
Paper proposes robot taxation as a policy response based on review of literature; presented as a policy recommendation rather than reporting new empirical estimation.
Therefore, accounting organizations in the region require targeted training, investment, and institutional support to improve AI adoption and use.
Authors' conclusion/recommendation based on survey and interview findings identifying skills and infrastructure gaps (not empirically tested within the study).
Perceived benefits of employing AI include increased efficiency, improved accuracy, better compliance, and more accurate decisions.
Self-reported perceptions collected via questionnaire and interviews, supported by thematic analysis (sample size not reported).
Accountants in Isabela (Region of Cagayan Valley) demonstrate strong analytical abilities.
Survey questionnaire and interviews analyzed using descriptive statistics and thematic analysis (sample size not reported in text).
GenAI-enabled education contributes to human capital development and organizational preparedness for AI-mediated workplaces, thereby contributing to business and management scholarship.
Thematic claim by the authors based on the review and conceptual integration; no direct empirical measurement provided within the review.
The review recommends practical implications for stakeholders (universities, business schools, firms, managers, policymakers), including curriculum redesign, responsible AI policy, AI literacy training, ethical assessment, and digital inclusion.
Author recommendations derived from the literature review and the conceptual framework (no empirical testing of recommendations reported).