Evidence (16496 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
Filter claims →
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 |
Apart from earnings adequacy, occupations characterized by dimensions of precarity were associated with lower LLM exposure (i.e., higher precarity on those dimensions corresponded to lower LLM exposure).
Abstract statement summarizing regression results across separate models for each precarity dimension (exact coefficients not provided in abstract).
Occupations most likely to be exposed to LLM are those where precariousness is lowest.
Summary conclusion based on the reported comparisons of mean LLM exposure across precarity categories using the Labour Force Survey and regression analyses described in methods.
Apart from earnings adequacy, LLM exposure was lower among occupations exhibiting each separate dimension of precarity (contractual instability, schedule unpredictability, working-time mismatch).
Separate multivariate linear regression models (one per precarity dimension) estimated associations between occupational LLM exposure and each dimension using Canada's Labour Force Survey; results reported in abstract (no per-dimension effect sizes provided in abstract).
Using the multidimensional precarity index, occupations characterized by low exposure to precarity had a significantly higher mean LLM exposure (mean 0.386, 95% confidence interval 0.356-0.417) compared to occupations with medium (mean 0.258, 95% CI 0.221-0.295), high (mean 0.260, 95% CI 0.194-0.328) or very high precarity (mean 0.205, 95% CI 0.136-0.275).
Analysis of Canada's Labour Force Survey; constructed multidimensional precarity index; multivariate linear regression models with cluster-robust standard errors; model coefficients used to produce mean estimates of occupational LLM exposure. (Sample size not reported in abstract.)
Algeria lags behind peer countries on key indicators of digital infrastructure, human capital, and institutional frameworks as evidenced by World Bank (2022) and Oxford Insights indices.
Specific comparative claim based on the paper's use of World Bank (2022) indicators and Oxford Insights Government AI Readiness Index scores; the summary does not report numeric index values or sample sizes.
Findings reveal that Algeria exhibits significant lag in digital infrastructure, human capital, and institutional frameworks compared to peers (Morocco, Egypt, Turkey).
Result reported from the paper's comparative analysis using World Bank indicators, the Oxford Insights Government AI Readiness Index, and sector-specific studies comparing Algeria to Morocco, Egypt, and Turkey; specific quantitative comparisons not provided in the summary.
Participant feedback attributes this vulnerability to minimal code review, plausible cover story, and overtrust in agents.
Qualitative analysis of participant feedback collected during/after the experiment; authors report these thematic attributions as explanations for the high failure-to-detect rate.
94% of developers fail to detect sabotage.
Reported quantitative result from the authors' user study with participants collaborating with the AI coding agents; percentage given in paper. (Sample described earlier as "Over 100 participants" but exact N for this result not stated here.)
Algorithmic scenario planning is being used for tax avoidance.
Presented in the abstract as an example of algorithmic technologies applied to international tax purposes (scenario planning for tax avoidance); no empirical details provided in the abstract.
Workers with a higher share of standardized routine tasks face more pronounced downward wage pressure.
Subgroup analysis by share of standardized routine tasks in workers' duties showing larger negative wage effects for those with higher routine-task shares.
The task substitution mechanism is the core channel underlying these effects of automation on wage structure.
Mediation/heterogeneity tests reported in the paper showing stronger automation effects where task substitution (standardized routine tasks) is higher; authors interpret this as the primary channel.
Wage growth for occupational groups with high exposure to automation lags markedly behind that of low-exposure groups.
Heterogeneity analysis across occupational exposure groups using CFPS panel data comparing wage growth trajectories for high- vs low-exposure occupations.
Existing research has significant shortcomings in terms of local empirical evidence, micro task mechanisms, and the impact of cutting-edge AI.
Critical appraisal in the paper's discussion of gaps identified through the systematic literature review; no single-study sample size.
Skill mismatch constitutes the core contradiction of labor force transformation.
Interpretive conclusion from the literature review asserting that mismatches between worker skills and job/task requirements are central to the labor-market effects of AI.
Despite the growing prevalence of human-AI decision making, the human-AI team’s decision performance often remains suboptimal, partially due to insufficient examination of humans’ own reasoning.
Motivating claim stated in the paper's introduction/abstract (appears to be based on broader literature and motivation rather than a new empirical test in this paper).
AACT also triggers higher cognitive load.
Reported measurement of cognitive load in the same house price prediction case study comparing AACT to traditional AI support (details and sample size not provided in abstract).
Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is 2.6%.
Empirical evaluation results reported by the authors summarizing ALE benchmark performance across mainstream harness and backbone configurations (no further detail on exact configurations or task/sample counts in excerpt).
The gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows.
Author argument presented in the paper; motivated by benchmarking limitations rather than an empirical test in the excerpt.
These gains have not translated into economically meaningful deployment across many professional domains.
Assertion in paper arguing a deployment gap between benchmark performance and real-world economic adoption; no quantitative deployment data provided in the excerpt.
Manual processing of these documents is time-consuming, inconsistent across reviewers, and unscalable.
Author claim / background motivation; no quantitative time or consistency metrics reported in the statement.
Actuaries rely primarily on structured numerical data for reserving and ratemaking, while valuable predictive information in unstructured text including medical records, adjuster notes, and call transcripts remains largely unused.
Author statement/observation in paper introduction; no empirical data or sample size provided to support prevalence claim.
The paper critically analyzes the implications of LLM-integrated search for brand trust, content authenticity, algorithmic bias, and market concentration.
Stated scope of analysis in the paper; presented as critical analysis rather than empirical claims in the provided text.
This paradigm shift raises critical questions regarding brand visibility, content authority, and digital marketing strategy.
Analytical claim in paper discussing implications; supported by theoretical argumentation rather than empirical measurement in the provided text.
The emergence of AI-generated summaries and answer-driven search experiences is shifting consumer discovery from link-based navigation to synthesized, context-aware responses.
Stated observation in the paper; argued via conceptual reasoning about AI-generated summaries and answer-driven interfaces rather than reported empirical metrics or sample-based experiments in the excerpt.
The condition 'prompt anxiety' describes a key feature of how stochastic systems organise cognitive labour under 'vector capitalism.'
Conceptual/theoretical framing introduced by the author to label and analyze user experience and labour organization; no empirical quantification provided in the abstract.
AI platforms transform this uncertainty into extractable value through subscription models, token-based pricing, and prompt marketplaces.
Political-economic / theoretical tracing in the paper citing platform business models (subscription, token pricing, prompt marketplaces) as mechanisms that monetize user uncertainty; no quantitative revenue or case-study sample sizes given in the abstract.
Analysis through LLMbench demonstrates that the uncertainty users experience corresponds to measurable variation in model confidence across the generated text.
Empirical demonstration using LLMbench visualisations (token probability distributions, entropy curves) to link user-reported uncertainty to measurable changes in model confidence; specific datasets, models, or sample sizes not provided in the abstract.
Users of large language models have to work with a measurably aleatory process: identical inputs produce different outputs and minor wording changes cascade through the probability field of the generated text.
Empirical analysis using the author's research instrument (LLMbench) for comparative close reading of LLM outputs; specific sample size or number of models/runs not reported in the abstract.
Prompt engineering resembles the psychological and temporal structures that Walter Benjamin identified in gambling behaviour.
Conceptual/theoretical argument presented in the paper drawing an analogy between prompt engineering practices and Walter Benjamin's analysis of gambling; no empirical sample size reported in the abstract.
Major risk pathways for agentic AI include hallucinations, prompt-injection attacks, autonomous decision errors, model drift, dependency failures, and cyber-physical harms.
Enumerative risk analysis within the paper summarizing plausible threat vectors and failure modes; based on theoretical reasoning and analogies to known AI and cyber risks rather than new empirical incident data.
These agentic-AI capabilities introduce novel exposures that do not fit neatly within traditional insurance categories such as cyber, professional liability, product liability, or directors and officers coverage.
Theoretical and market-structure analysis in the paper comparing agentic-AI exposures to existing insurance lines; illustrative examples and taxonomy rather than quantified empirical tests.
Agentic artificial intelligence (AI) systems are transforming the risk landscape by extending beyond information generation to autonomous planning, tool invocation, decision execution, and persistent modification of digital and physical environments.
The paper's conceptual argument and framing/abstract describing agentic AI capabilities and their implications; theoretical analysis rather than empirical measurement.
Collaborative filtering and graph-based recommendation models are highly effective because they leverage observed user interactions, but this dependence creates a fundamental cold-start challenge when newly added content has no interaction history.
Statement in paper framing problem; references to general properties of collaborative filtering and graph-based recommenders (conceptual / literature-backed claim, no specific experiment reported in this excerpt).
The determining barrier to adoption observed in the two studied public-service units was not technological but training-related.
Qualitative analysis and intervention observations across two auditable case studies (SES/CONT in 2024 and UCI/SEDET in 2025); author-developed intervention and outcome changes used to support inference.
The adoption of generative artificial intelligence in the public sector has been treated predominantly as a technological problem, with the expectation that productivity gains would follow from more capable models.
Author statement / literature-positioning in paper (assertion about prevailing treatment); no quantitative data provided in text to support prevalence.
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).
AI-based career planning platforms and digital portfolio/performance trackers can embed biases, amplify pressures for self-optimisation, provide only generic recommendations, and risk promoting a narrow view of what constitutes a desirable career.
Conceptual concerns and literature cited in the editorial (Bankins et al., 2024a and other referenced works); argued as potential unintended consequences rather than direct evidence from a single large empirical study.
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.
STARA is displacing routine tasks and potentially entire roles, particularly in occupations where automation and robotics can substitute standardized work processes.
Synthesis of existing literature cited in the editorial (e.g. Bahadure et al., 2024; Oosthuizen, 2019, 2022; Singh and Chandra, 2026; Singh et al., 2026).
By framing AI risk exclusively in cybersecurity terms, the Order constructs an AI-risk universe in which provenance, labor, education, culture, meaning, and the commons are rendered 'not testable' within the policy regime.
Argumentative/theoretical claim backed by textual analysis and the counted absence of relevant terms in the EO.
The Executive Order frames AI risk overwhelmingly through cybersecurity language.
Textual analysis of the EO; supported by the paper's verified word-count analysis showing high frequency of security/cyber terms relative to other domains.
The COVID-19 pandemic reduced tourism’s GDP share by approximately 37%.
Fixed-effects panel estimation including a COVID-19 indicator on 33 countries (2017–2023); reported coefficient β = –0.455, p < 0.001 (interpreted as ~37% reduction in the dependent variable).
AI adoption intensifies existing sustainability challenges for the newsroom, as journalistic content and labour increasingly support AI systems without corresponding financial return.
Qualitative interview data and organisational analysis from Al-Masry Al-Youm indicating increased use of journalistic outputs for AI purposes and lack of matched revenue; sample size not reported in the excerpt.
Reliance on global technology providers embeds forms of platform dependency within newsroom operations at Al-Masry Al-Youm.
Qualitative case study based on in-depth interviews with journalists, editors, and technical staff at Al-Masry Al-Youm (Egypt); analysis of newsroom practices and integration of third-party/global AI tools. Sample size not reported in the excerpt.
An incentive sweep reveals Goodhart-style drift where measured performance becomes anti-correlated with true outcomes.
Simulation results in Medi-Sim showing that optimizing measured metrics leads to a decrease (anti-correlation) in true outcomes (Goodhart effect).
Existing healthcare AI benchmarks hold this [strategic provider] response fixed and so cannot evaluate mechanisms by the equilibrium they produce.
Author statement/argument in the paper about limitations of existing benchmarks (conceptual claim; not an empirical experiment).
Research on platform governance remains fragmented and lacks an integrative perspective.
Conclusion drawn from the systematic literature review (644 publications) indicating fragmentation in the scholarly literature.
Participants in platform ecosystems cannot be governed through traditional command-and-control mechanisms.
Conceptual claim supported by the literature synthesized in the systematic literature review (644 publications).
Government subsidies exert a negative moderating influence on the relationship between fintech development and corporate total factor productivity.
Moderation analysis reported in the paper on Chinese A-share listed manufacturing firms (2015–2023); paper states government subsidies weaken the positive fintech–TFP relationship (no numeric interaction estimates provided in the excerpt).
Research on AI-enabled decision-making and upper echelons theory (UET) has largely evolved in parallel (i.e., the two literatures are not well integrated).
Concept-centric literature review mapping management and IS literatures and identifying lack of integration (no quantitative meta-analysis or sample size reported).