Evidence (198 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 |
We find marked heterogeneity in model predictions.
Empirical comparison of multiple projection models (comparison of outputs across the set of models described in the paper).
Youthful populations, large shares of informal and agricultural employment, and concentrated digital and outsourcing hubs create a patchwork of exposure: pockets of intense AI adoption sit beside vast swathes of low-digitisation employment.
Conceptual synthesis and descriptive evidence cited from major international organizations and industry studies (World Bank, ILO, etc.) in the paper; no numerical sample provided.
Both full-panel estimates (hiring and productivity) are imprecise, pointing to augmentation rather than displacement.
Author discussion noting imprecision of full-panel coefficients and interpreting the pattern as consistent with augmentation (productivity up, hiring down slightly but imprecise).
AI exposure varies sharply across Indian firms.
Author statement based on firm-level exposure measures computed in the paper (weighted averages of occupational LLM exposure scores across business lines).
The dominant paradigm has shifted from 'substitution' (machines replacing workers) to 'augmentation' (AI augmenting human work).
Interpretive conclusion in the paper drawn from secondary literature (WEF, ILO, McKinsey, PwC) and observed policy/industry trends.
AI exposure is more positive for occupations performing nonroutine interactive work and more negative for occupations concentrated in analytical, scientific, and operations-control skills.
Occupation-level analysis mapping skill content (interaction-and-communication vs. analytical/scientific/operations-control) to market-implied AI premium; comparison across occupational skill categories.
There is a growing reliance on agentic AI systems within the platform context.
Qualitative evidence from the 20 interviews and the 24-participant workshop reporting increased dependence on AI agents for tasks and decision support.
There is increasing automation of operational tasks in the development domain.
Participant reports and workshop discussions from 20 interviews and a 24-person workshop indicating automation of operational activities; qualitative thematic evidence.
The intended contribution is an Information Systems framework explaining when AI supports human augmentation and when it produces functional substitution.
Stated intended theoretical contribution in the abstract (proposed framework). This is an intended outcome rather than an empirically demonstrated result in the provided text.
The present wave of automation targets non-routine cognitive activity such as coding, technical writing, and graphic design, unlike past automation which mainly involved routine manual activity.
Framing/background statement in the paper contrasting historical automation (routine manual tasks) with current AI-driven automation of non-routine cognitive tasks; no sample size or quantitative test reported in the abstract.
AI's rapid evolution has profound effects on the labor market, influencing the levels, skills needed for jobs, and overall jobs content.
Statement from the paper's synthesis/introduction summarizing reviewed empirical studies (systematic literature review covering studies from 2017–2025). Number of underlying studies not reported in the excerpt.
Although the geometry (bipolar structure) is stable, its content is not: across a decade the polarity has inverted relative to Frey and Osborne (2013).
Comparison of macro-level placements between the paper's LLM-era OAI and the Frey-Osborne (2013) rankings; authors report inversion and supporting correlation statistics.
Tool-Mediated Physical (M2) and Planning & Design (M7) are separated by Cohen's d = 2.41 (H = 172.88, p = 6.21e-34).
Statistical comparison reported in the paper (Cohen's d, H-statistic, p-value) between the two macro clusters' OAI distributions.
Projecting the DWA-level Occupational Automation Index (OAI) onto a 7-macro semantic typology produces a bipolar structure (two poles separated by a low-contrast middle band).
Authors' projection of previously computed DWA-level OAI onto a 7-cluster semantic typology and subsequent analysis of cluster structure.
Artificial Intelligence (AI) has changed how people work across various fields and businesses, especially in the Indian Information Technology (IT) industry.
Authors' qualitative synthesis of peer-reviewed literature and thematic evaluation of secondary data (literature review). No sample size reported.
Generative AI exposure is dynamic rather than fixed, changing substantially over time.
Empirical application of the dynamic posting-level exposure measure to the nationwide job-postings data showing substantial temporal change (as stated in the paper's findings).
AI embeds algorithmic actors into the microfoundations of strategy, altering the role and behavior of individual-level actors that underlie firm-level phenomena.
Conceptual analysis of Microfoundations literature; theoretical proposition that algorithms act as actors at micro levels; no empirical sample provided.
Artificial intelligence (AI) is rapidly reshaping knowledge-intensive work by automating, augmenting, and reconfiguring core professional activities.
Paper asserts this as a motivating observation based on prior literature and descriptive claims; no original empirical sample or quantified data reported.
Current models about the vulnerability level of occupations and economic sectors differ widely in their forecasts.
Paper's comparative statement about existing models and their forecasts (no specific models, quantitative comparisons, or sample sizes provided in the excerpt).
Agents are not labor; they are a production technology that converts compute capital K_c into effective units of cognitive labor L_A.
Theoretical argument and definitional framing in the paper: the authors recast agents as a technology that transforms compute capital into effective cognitive labor units within an analytical model (textual/theoretical exposition). No empirical sample or experimental data reported in the excerpt.
High-value uses require broader authority exposure — data access, workflow integration, and delegated authority — when governance controls have not yet decoupled capability from authority exposure.
Conceptual/mechanism claim articulated in the paper (motivating assumption for the analytical model; no empirical sample given in the abstract).
Experienced developers maintain control through detailed delegation while novices struggle between over-reliance and cautious avoidance.
Observed behaviors and accounts from the AI-assisted debugging task (10 juniors) and senior participants in ACTA/Delphi and blind review phases (5 + 5 seniors).
Frontier technologies remain concentrated in specialised occupations, while digital technologies are widespread.
Distributional analysis of OTSS across occupations showing concentration patterns of frontier technologies versus ubiquity of digital technologies.
For the average worker in 2023, manual technologies account for the largest share of skill content (42 per cent), followed by digital (38 per cent) and frontier technologies (20 per cent).
Computed OTSS applied to occupation-level data for Germany in 2023; reported shares for the "average worker".
Bounded agents act as an amplifying but not necessary extension to the foundation-model stack for changing work coordination.
Conceptual argument within the paper distinguishing bounded agents from the core stack; no empirical comparison or measurement reported.
Mathematics (SAFI: 73.2) and Programming (71.8) receive the highest automation feasibility scores; Active Listening (42.2) and Reading Comprehension (45.5) receive the lowest.
SAFI benchmark results reported for specific O*NET skills (numerical SAFI scores provided in the paper).
The central tension in AI for science is between automation (building systems that replace human researchers) and augmentation (tools that amplify human creativity and judgement).
Analytical claim based on the paper's review of historical examples and conceptual discussion; no primary data or experimental design reported.
Science has repeatedly delegated its bottlenecks to machines—first inference, then search, then measurement, then the full workflow—and each delegation solves one problem while exposing a harder one underneath.
Interpretive historical argument drawing on examples across AI-for-science milestones (e.g., DENDRAL, search and inference systems, measurement automation, and contemporary end-to-end workflows). No quantitative sample or experimental method reported.
The frontier is jagged: humans retain decisive advantages in long-horizon reliability, genuinely novel problems, calibrated self-knowledge, sample-efficient learning, and embodied action.
Synthesized comparison between AI benchmark crossings and remaining capability gaps documented in the paper (qualitative summary of areas where humans still outperform AI); no specific empirical studies, sample sizes, or effect estimates provided in the excerpt.
A closed-form automation-debt measure (ρ(P)) formalises how role-level decisions accumulate across multi-step processes, and its warning is neutralised only by a regulator-mandated human-in-the-loop anchor.
Analytic/formal model presented in the paper (closed-form measure and its properties); theoretical derivation described in the abstract.
India’s large IT–BPO workforce is flagged as highly susceptible to automation by 2030.
Statement in the paper referencing projections and sectoral vulnerability drawn from international forecasts and industry analyses (no numeric susceptibility metric provided).
Formal, export-oriented, and digitally intensive sub-sectors—IT–BPO, ready-made garments (RMG), customer support, and banking—carry the highest AI exposure.
Empirical pattern reported in the paper based on quantitative syntheses from institutions (World Bank, ILO, WEF) and industry studies; no sample size or numerical exposure metric provided.
India has low aggregate AI exposure, ranking 123rd of 149 countries.
Author statement reporting a cross-country ranking (presumably from an AI-exposure index referenced in the paper).
Women in high-income countries face a risk of automation nearly three times higher than men due to their concentration in administrative roles.
Paper's secondary quantitative synthesis attributing a ~3x relative risk to occupational gender segregation (administrative roles); based on international report data referenced in the study.
At the micro level, elevated risk salience related to privacy, safety, or ethical concerns may lead users to adopt guarded interaction strategies characterized by reduced contextual disclosure and limited iteration.
Theoretical proposition within the paper's guarded engagement loop framework, drawing on prior research in privacy calculus and algorithm aversion; no specific empirical data reported in the abstract.
Expertise moderated the effect of LLM guidance: novices exhibited passive AI reliance.
Stratified analyses by participant expertise level using behavioral and eye-tracking measures indicating novices shifted attention to the AI/chat and exhibited more passive acceptance of guidance.
The rapid emergence of AI agents as intermediaries between humans and web content invalidates the web's human-first assumption.
Paper's conceptual claim based on observed/assumed rise of AI agents acting as intermediaries; no quantitative data or sample presented in provided text.
In the production stage, workers are alienated into becoming data producers.
Conceptual claim based on Marxian analysis of labor and data extraction; no empirical sample or quantitative evidence presented.
One in three Scheduled Tribe (ST) graduates work in farm or elementary occupations untouched by AI.
Occupational distribution from PLFS 2025 after mapping AI-exposure indices; reported share of ST graduates in farm/elementary (AI-unexposed) occupations in the 83,000-employed-graduate sample.
One in four Scheduled Caste (SC) graduates work in farm or elementary occupations untouched by AI.
Occupational distribution from PLFS 2025 after mapping AI-exposure indices; reported share of SC graduates in farm/elementary (AI-unexposed) occupations in the 83,000-employed-graduate sample.
Graduates from the Scheduled Castes and the Scheduled Tribes are 0.24--0.37 standard deviations less exposed than upper-caste graduates within the same district.
Within-district comparisons using three occupational AI-exposure indices mapped to PLFS 2025; reported standardized exposure differences for SC and ST graduates relative to upper-caste graduates in the 83,000-employed-graduate sample.
Macro-level correlation between Frey-Osborne (2013) and Eloundou-era rankings is Spearman rho = -0.750, p = 0.020 (against the original Oxford Martin appendix), indicating inversion.
Reported Spearman correlation and p-value comparing macro-level rankings between the original Frey-Osborne appendix and the paper's Eloundou-era results.
Tool-Mediated Physical (macro M2) has mean OAI = 0.054.
Reported macro-level mean OAI computed after projecting DWA OAI values into the 7-macro typology.
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.)
AI disproportionately affects routine and mid-skilled jobs.
Synthesis of literature (2010–2024) reported by the authors indicating disproportionate automation/AI exposure for routine and mid-skilled occupations.
Women are overrepresented in occupations predicted to be more affected by generative AI (using pre-ChatGPT occupational sorting).
Descriptive analysis of Swedish administrative data characterizing occupational gender composition before the release of ChatGPT and mapping occupations to predicted exposure to generative AI.
Gig workers, though formally classified as independent contractors, are functionally subjected to pricing control, performance monitoring, automated penalties, and deactivation mechanisms that closely resemble managerial authority.
Descriptive/qualitative evidence in the paper: examples and analysis of platform design and management practices (algorithmic pricing, monitoring, penalties, deactivation); based on platform policy documents, case examples and comparative review (no quantitative sample size reported).