Evidence (484 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
20058 claims
Filter claims →
Productivity
17184 claims
Filter claims →
Governance
16099 claims
Filter claims →
Human-AI Collaboration
16034 claims
Filter claims →
Innovation
10501 claims
Filter claims →
Org Design
10496 claims
Filter claims →
Labor Markets
6444 claims
Filter claims →
Skills & Training
5385 claims
Filter claims →
Inequality
4148 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 | 1820 | 479 | 278 | 1820 | 4588 |
| Organizational Efficiency | 2711 | 616 | 401 | 173 | 3922 |
| Governance & Regulation | 2075 | 886 | 459 | 246 | 3714 |
| Technology Adoption Rate | 1467 | 530 | 258 | 206 | 2488 |
| Decision Quality | 1281 | 496 | 289 | 152 | 2228 |
| Output Quality | 1227 | 447 | 207 | 138 | 2025 |
| AI Safety & Ethics | 634 | 754 | 207 | 83 | 1688 |
| Research Productivity | 826 | 241 | 114 | 422 | 1624 |
| Firm Productivity | 1052 | 154 | 163 | 66 | 1441 |
| Task Allocation | 685 | 211 | 331 | 99 | 1335 |
| Market Structure | 433 | 423 | 242 | 46 | 1150 |
| Innovation Output | 639 | 91 | 105 | 34 | 871 |
| Task Completion Time | 476 | 113 | 43 | 36 | 672 |
| Firm Revenue | 445 | 126 | 58 | 25 | 656 |
| Skill Acquisition | 364 | 119 | 109 | 34 | 626 |
| Consumer Welfare | 288 | 167 | 104 | 31 | 592 |
| Employment Level | 214 | 140 | 174 | 50 | 582 |
| Error Rate | 230 | 251 | 35 | 16 | 535 |
| Fiscal & Macroeconomic | 268 | 136 | 71 | 50 | 532 |
| Inequality Measures | 100 | 307 | 96 | 12 | 515 |
| Worker Satisfaction | 221 | 173 | 60 | 30 | 484 |
| Automation Exposure | 155 | 138 | 65 | 36 | 398 |
| Regulatory Compliance | 171 | 120 | 30 | 13 | 335 |
| Developer Productivity | 222 | 58 | 27 | 13 | 321 |
| Team Performance | 188 | 56 | 50 | 24 | 320 |
| Wages & Compensation | 146 | 104 | 46 | 16 | 312 |
| Training Effectiveness | 207 | 41 | 21 | 26 | 298 |
| Job Displacement | 23 | 153 | 52 | 4 | 232 |
| Hiring & Recruitment | 102 | 57 | 30 | 11 | 202 |
| Skill Obsolescence | 16 | 102 | 24 | 6 | 148 |
| Creative Output | 71 | 42 | 23 | 6 | 143 |
| Social Protection | 57 | 30 | 11 | 3 | 101 |
| Labor Share of Income | 29 | 42 | 24 | 2 | 97 |
| Worker Turnover | 43 | 29 | 6 | 4 | 82 |
| Industry | — | — | — | 1 | 1 |
Employee resistance to AI adoption is driven by fears of job displacement and skill obsolescence, and the paper argues that training programmes and transparent change management can help overcome this resistance.
Thematic synthesis and practical recommendations in the systematic literature review; the supplied text reports no employee survey sample or estimated intervention effect.
Digital visibility has dual and divergent effects: it can increase felt recognition and relational connection while also creating performative pressures that weaken authentic belonging.
Conceptual theorization grounded in relational cohesion theory and sociomaterial perspectives; supported by literature synthesis rather than primary empirical data.
The two ontologies imply different process explanations for organizational outcomes: correspondence models emphasize trait alignment leading to outcomes such as satisfaction and retention, whereas constitutive models emphasize interpretive work leading to meaningfulness and adjustment.
The article develops separate processual explanations for each ontology; the claim is conceptual rather than an estimate from observed data.
Across three experiments, AI emotional support was rated as superior to human support only among participants who had initially chosen AI.
Three experiments with a combined initial sample of 1,951 participants; after preregistered exclusions, 1,894 contributed to pre-conversation analyses, and Studies 2–3 included randomized assignment to congruent or incongruent human/AI partners.
Indian banking employees report both positive views of AI's contribution to accounting, sales, contract management, and cybersecurity and concerns about job security, alongside demands for structured employer-provided training and upskilling support.
The paper summarizes a mixed-methods study by Dwivedi and Kochhar (2023) involving surveys and interviews with Indian banking employees; the sample size is not stated in the supplied text.
Algorithmic management has contested effects on employee well-being: reduced autonomy and opaque decision logic are associated with diminished job satisfaction and well-being, while transparent design and algorithm-supported autonomy can improve fairness perceptions and job satisfaction.
Synthesis of studies examining autonomy, transparency, surveillance intensity, perceived fairness, and employee well-being under algorithmic management.
The study found that the perceived or disclosed source of performance feedback influences how employees evaluate the feedback.
A vignette-based randomized experiment varied only the disclosed feedback-production workflow while presenting all participants with the same feedback text.
Hybrid feedback generally produced reactions between those elicited by fully human-written and fully AI-generated feedback.
The experiment compared fully human-written, fully AI-generated, AI-generated and human-refined, and human-written and AI-refined feedback conditions while holding the feedback text constant.
AI-based performance feedback and automated monitoring can either enhance or undermine employee engagement by changing perceived job demands, resources, and psychological contracts.
Conceptual extension of JD-R and Psychological Contract Theory to algorithmic management; no AI implementation sample or causal estimate is reported.
Perceived fulfillment or breach of the psychological contract changes employee trust and engagement.
Conceptual synthesis of Psychological Contract Theory and employee-engagement mechanisms; the paper does not report a primary empirical estimate.
Job resources such as support, autonomy, and feedback increase employee engagement, whereas excessive job demands can reduce engagement; challenging demands may sometimes be motivating.
Conceptual application of the Job Demands-Resources model in the literature synthesis; no reported effect estimate or primary sample.
The paper proposes Dynamic Capability-Dependency Theory (DCDT), in which AI-mediated well-being operates across two time horizons: immediate affective benefits and longer-term changes in capability, autonomy, relatedness, meaning, and dependency.
Theoretical framework and dynamic state model distinguishing short-term experienced well-being from accumulated stocks that evolve with repeated AI use.
Workers' perceptions of psychological contract fulfillment are contingent on algorithmic transparency, interpretability, and perceived justice.
Thematic synthesis of 39 empirical studies examining algorithmic management and psychological contract or social exchange outcomes.
Users of the organization’s internal chatbot rated quality and factuality as the most important dimensions, while sustainability and inclusion received substantially lower priority.
The authors surveyed 429 users of the organization’s internal chat assistant using questions similar to those given to advisory-board members.
AI companions can reduce momentary loneliness in controlled and short-term settings, but observational evidence indicates that outcomes vary according to loneliness and existing social connectedness.
The paper summarizes De Freitas et al. (2025) on controlled short-term experiments and Nakagomi et al. (2026) on observational evidence.
Digital Taylorism creates a paradox in which workers may receive flexibility while facing significant digital control and accountability pressures.
Literature synthesis concerning gig work, algorithmic asymmetry, flexibility, surveillance, and performance pressures.
Autonomous systems may perform measurable technical tasks better than humans while leaving a human vocational residue involving trust, accountable judgment, conversation, and choice.
Argument based on the hypothetical example of a robotic cutting table outperforming human cutting accuracy while clients continue returning for interpersonal consultation.
Plainly including sustainability information improved users’ subjective evaluations of the recommendation system, but this improvement did not translate into greener choices.
The randomized studies measured both selection behavior and subjective evaluations across explanation conditions.
The study identifies distinct emotional and decision-making dynamics before, during, and after a career break.
Reflexive thematic analysis organized the findings into three phases: before the break, during the break, and after returning to work.
Women took career breaks for both caregiving or other familial obligations and deliberate self-care, including recuperation and rebuilding mental and physical health.
Thematic analysis of accounts from 22 urban Indian women in IT who took career breaks and subsequently returned to work.
Augmentation enabled by AI agents will be conditional and fragile (augmentation with fragile conditions).
Argument/analysis in the paper describing conditions required for successful augmentation and potential fragility; no empirical evidence provided.
The mediating role of team role overload in the relationship between leader–AI collaboration and team performance showed mixed results.
Authors report inconsistent/mixed empirical findings for team role overload as a mediator across the two multi-source, multi-wave surveys (no sample size or statistics provided in abstract).
Punishment controllability moderates the relationship between punishment severity and algorithm aversion.
Reported empirical moderation test in the study (methodological details and sample size not provided in the abstract).
Leadership styles, organizational culture, employee proficiency, and engagement measures are some of the factors that affect psychological well-being.
Literature review identifying correlates / determinants of psychological well-being at work; based on synthesized studies rather than a single primary dataset in the excerpt.
The algorithm did not eliminate all inequities but made decision constraints visible and debatable.
Findings reported from administrative data, ethnographic observations, internal documents, and informal feedback collected in the single-hospital case study.
Fluent explanations systematically increase user confidence and reliance on AI without reliably improving, and in some cases undermining, task accuracy (the 'Persuasion Paradox').
Stated summary claim based on results across three controlled human-subject studies described in the paper (multi-stage reveal design; between-subjects comparisons). Specific sample sizes not reported in the provided text.
Effects of digital HR interventions are non-linear and context-dependent, moderated by privacy, explainability, digital literacy, and regulation.
Cross-study synthesis reporting moderation and context-dependency across the 47 included studies; thematic identification of moderating factors.
AI may greatly improve the work-life balance only when it is supported by a positive culture, a set of rules, sufficient training, and protection against excess workload.
Overall conclusion of the review article synthesizing mixed findings across studies; conditional/qualified statement rather than a quantified causal estimate.
There was inconsistent mediation: the direct effect of explanation quality on trust, controlling for fairness, was negative (b = −0.81, p < .001).
Mediation analysis reported from the experiment (N = 275); paper reports direct effect coefficient b = −0.81 with p < .001 when controlling for fairness.
Organisational culture shapes employees' concealment behaviours around AI use.
Paper asserts that organisational culture influences these behaviours; no methodological details or sample size are given in the provided text.
Extending technostress theory from digital tools to factory automation, shifting performance norms and cognitive burdens generate mental spillovers beyond the shop floor.
Paper frames empirical findings as consistent with an extension of technostress theory and interprets patterns (timing and dimensions of effects) as evidence of shifting performance norms and cognitive burdens producing mental spillovers.
Qualitative interviews show that employees view the twin (green and digital) transition in diverse ways.
Findings drawn from over 100 qualitative interviews in a leading OEM, reported in the paper.
Conflicting reactions and feelings among employees were prevalent alongside productivity gains and concerns about employment preservation
Qualitative analysis and sentiment analysis applied to employee-related data (methods: sentiment analysis, thematic evidence); textual/qualitative evidence reported alongside quantitative productivity findings
A clear gap emerged between structural prominence and emotional intensity: riders discussed time pressure most widely and largely accepted it as routine, while punitive mechanisms and health-protection deficits drew the strongest negative responses.
Comparison of topic prevalence (coverage) and sentiment intensity from 10,103 comments supplemented by interview quotes showing prevalence vs emotional valence differences.
Practitioners sharply disagree about how coding agents change code review: whether review becomes the bottleneck, whether human review remains necessary, and whether agents erode the understanding that review once built.
Synthesis of practitioner discourse at scale via collected grey-literature (engineering blogs and Reddit threads) and a coded sample; claim summarizes observed disagreement in practitioner sources.
These patterns suggest a commoditization effect of AI on labor, with implications for online labor market design, workers' incentives to invest in human capital, and labor welfare.
Interpretation synthesized from the three empirical findings above (decline in human-capital importance, rise in price importance, decline in demand premium for high-human-capital workers, and reallocation toward lower-priced workers). This is presented as the paper's conceptual/mechanistic conclusion and policy implication rather than a separately tested causal estimate. (Empirical basis: Upwork analysis and difference-in-differences; sample size not reported in abstract.)
Dominant comments shifted in tone from mockery toward gatekeeping and structural protest.
Speech-act coding of 300 confirmed accusations and sentiment/trajectory analysis showing relative decline in mockery-coded acts and increase in gatekeeping/structural-protest acts over time.
There is a significant U-shaped relationship between AI application and employees' job insecurity: moderate AI application reduces insecurity, whereas excessive application heightens it.
Empirical analysis of cross-sectional self-reported questionnaire data collected from employees (411 valid responses) using regression-type analyses reported as showing a significant U-shaped relationship between AI application intensity and job insecurity.
This research employed a vignette experiment to investigate how the embeddedness of GenAI and organizational authorization impact employees' negative emotion (specifically guilt) and risk perception.
Stated method in paper: a vignette experiment was used to test effects on guilt and risk perception. (No sample size reported in the provided text.)
Algorithmic systems for productivity and performance monitoring generate efficiencies but also create new pressures in technology-mediated work environments, including the tracking of employees’ emotional and physiological responses at work and during non-work time.
Literature synthesis and citations (e.g. Giermindl et al., 2022; McCartney and Fu, 2022; Norlander et al., 2021; Downie et al., 2025).
Acquiescent silence (resignation-based) is motivationally distinct from defensive (fear-driven) silence.
Theoretical distinction advanced using organisational silence literature (conceptual claim referencing existing theory).
AI opacity, automation intensity, anthropomorphic and affective design features, and the degree of human-centered system design are determinant factors shaping users' psychological responses to human–AI collaboration.
Authors' synthesis from reviewed empirical and theoretical studies highlighting design and system characteristics associated with psychological outcomes.
The interdisciplinary literature identifies technostress, automation fatigue, cognitive overload, algorithmic anxiety, overtrust, and responsibility ambiguity as key phenomena arising from integration of AI systems and AI-enabled robots into collaborative human work environments.
Synthesis of interdisciplinary peer-reviewed studies (systematic review); topics extracted from reviewed papers as reported by the authors.
The reviewer-effort collapse creates a welfare misalignment: authors benefit from a weakened 'rat race' while editors suffer from degraded signal informativeness.
Comparative statics and welfare analysis in the theoretical model showing authors' equilibrium payoffs rise as competition/polishing dissipates, while editor's signal informativeness declines due to lower reviewer effort.
Social, cultural, and ethical considerations influence women’s engagement in AI-centric workplaces.
Claim made in the review, based on interdisciplinary literature that includes sociocultural analyses and ethical discussions; the abstract does not provide empirical effect estimates or sample sizes.
Sensitivity analyses indicate the observed positive belief changes likely reflect recovery from carry-over effects rather than genuine training-induced shifts.
Authors' sensitivity analyses discussed in the paper that examined alternative explanations (e.g., carry-over effects) and concluded the belief-change result is likely due to recovery from such effects.
Qualitative results underscored both perceived benefits in comprehension and challenges when interpretations of gaze behaviors were inaccurate.
Qualitative analysis of participant feedback from the study (n=36) reporting themes of improved comprehension and occasional problems when the assistant misinterpreted gaze.
AI adoption has an inverted U-shaped effect on employee-related corporate social responsibility (ECSR).
Panel regression with quadratic specification (AI and AI^2) showing statistically significant positive coefficient on AI and statistically significant negative coefficient on AI^2; sample of 2,575 Chinese listed firms observed 2013–2023; controls, firm and/or year fixed effects and robustness checks reported.
Digital visibility can generate anxiety and stress, reduce perceived privacy, and weaken psychological safety.
Theoretical emotional mechanism in a conceptual model derived from literature synthesis; not an empirically estimated effect.
Digital visibility can foster simulated belonging through outward displays of connection or availability, such as online status, rapid replies, and curated digital presence, without corresponding relational substance.
Conceptual distinction between authentic and simulated belonging developed through theorization of visibility-related behaviors.