Evidence (4137 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 |
Governance
Remove filter
Sustain investment in shared, standardized data infrastructures (datasets, ontologies, benchmarks) to support medical algorithm–hardware co-design.
Workshop infrastructure call presented during breakout sessions and final recommendations at the NSF workshop (Sept 26–27, 2024).
Principal recommendation: shift from isolated algorithm or hardware efforts to integrated algorithm–hardware–workflow co-design for medical contexts.
Stated workshop recommendation derived from panels and cross-disciplinary consensus at the NSF workshop (Sept 26–27, 2024).
Sustained public investment and new validation, governance, and translation ecosystems are needed to de-risk commercialization and accelerate safe, accountable clinical adoption.
Workshop principal recommendation based on qualitative synthesis of expert judgment from participants and breakout outcomes (NSF workshop, Sept 26–27, 2024).
Enabling next-generation medical technologies requires a fundamental reorientation toward algorithm–hardware co-design that is clinic-aware, validated continuously, and backed by shared data and compute infrastructures.
Consensus recommendation from a two-day NSF workshop (Sept 26–27, 2024) in Pittsburgh convening interdisciplinary participants (academic researchers in algorithms and hardware, clinicians, industry leaders). Methods: expert panels, thematic breakout sessions, cross-disciplinary discussions, consensus-building. Documentation at https://sites.google.com/view/nsfworkshop.
RAT data could be valuable for training models that better emulate human interpretive processes; firms owning such data may gain competitive advantage.
Argument in the AI economics section; no empirical model-training experiments or market analyses provided.
RATs make readable and potentially quantifiable the preparatory interpretive work that contributes to downstream outputs, with implications for labor accounting and human capital valuation.
Theoretical economic and policy discussion in the paper; no empirical measurement or case studies provided to quantify how much preparatory work is captured or its economic value.
RATs can enable collective sensemaking via shared trails and networked associations among readers.
Conceptual argument and suggested network-analysis methods; illustrated with the speculative WikiRAT use case. No group-level empirical studies reported.
RATs can support richer reader models (personalization and modeling of interpretive behavior) through sequence analysis, embedding/clustering of trajectories, and other analytic techniques.
Proposed analytical methods (sequence analysis, embedding/clustering, network analysis) listed in the paper; no implementation results or quantitative evaluations provided.
RATs enable reflective practice by helping readers see and revise their own processes.
Proposed affordance in the paper based on the inspectable nature of RATs and the WikiRAT illustration; suggested as a potential use case rather than empirically demonstrated.
RATs treat reading as a dual kind of creation: (a) creative input work that shapes future artifacts, and (b) a form of creation whose traces are valuable artifacts themselves.
Theoretical proposal and design rationale presented in the paper; illustrated via a speculative prototype (WikiRAT). No empirical validation provided.
Reading Activity Traces (RATs) reconceptualize reading — including navigation, interpretation, and curation across interconnected sources — as creative labor.
Conceptual argument in the paper; supported by theoretical framing and literature review rather than empirical data. No sample size or deployment reported.
Structural fixes — altering environment design or policy class to ensure the induced Markov chain is ergodic (e.g., ensuring mixing/recurrence or preventing absorbing bad states) — can eliminate the ensemble/time-average gap.
Paper discussion and examples suggesting interventions to change chain structure; conceptual/theoretical proposal supported by illustrative examples (no empirical deployment studies).
Robust/adversarial and model-uncertainty methods can hedge against trajectories that lead to poor long-run behavior and thus mitigate risks from non-ergodic dynamics.
Survey of robust control and adversarial RL approaches in the paper; conceptual argument linking robustness to protection against adverse sample paths; no new empirical tests.
Ergodic control and sample-path optimality formulations recast control objectives in terms of time averages or almost-sure sample-path criteria rather than ensemble expectations and are therefore appropriate for single-trajectory performance targets.
Survey and formal discussion in the paper connecting ergodic control literature to single-trajectory objectives; theoretical references summarized.
Almost-sure and probabilistic constraint methods (chance constraints, safe RL) can enforce that long-run performance exceeds thresholds with high probability, addressing single-trajectory guarantees.
Surveyed methodologies and references in the paper describing chance-constrained and safe RL formulations; conceptual synthesis rather than empirical demonstration.
Distributional reinforcement learning (optimizing the full return distribution) enables optimizing objectives such as median, lower quantiles, or CVaR which better reflect single-run guarantees.
Literature survey in the paper citing distributional RL approaches and linking them conceptually to single-trajectory guarantees; no new experiments provided.
Risk-sensitive and utility-based objectives (e.g., maximize expected utility such as log-utility or minimize downside risk) can produce policies that prefer more reliable time-average outcomes compared to raw expected-reward objectives.
Surveyed literature in the paper summarizing risk-sensitive and utility-based RL approaches; conceptual argument rather than new empirical validation.
Numerical simulations confirm the analytic extreme-value scaling for earliest discoveries and demonstrate that introducing non-reciprocal biases leads to stable monopolies whereas symmetric interactions do not.
Numerical simulations (stochastic realizations) reported in the paper used to validate analytic predictions and illustrate dynamical outcomes; however, the summary does not specify simulation sample sizes, parameter sweeps, or robustness checks.
Empirically, RAD improves out-of-distribution (OOD) robustness (OOD harmlessness) compared to baselines.
Out-of-distribution harmlessness evaluations reported in the paper showing RAD performs better than baselines on OOD safety tests (exact experimental details not provided in the summary).
Empirically, RAD improves harmlessness relative to baseline RLHF methods.
Empirical evaluations reported in the paper comparing RAD to baseline RLHF methods on harmlessness metrics (specific datasets, sample sizes, and exact metrics not provided in the summary).
Entropic regularization plus Sinkhorn iterations yields a differentiable, computationally tractable objective suitable for end-to-end optimization with policy gradient methods.
Algorithmic design and implementation details in the paper showing use of entropic-regularized OT and Sinkhorn; claimed compatibility with policy-gradient/end-to-end training (no concrete runtime benchmarks or sample-complexity numbers in the summary).
Policy recommendations: economists and policymakers should perform cost–benefit analyses of explainability mandates, incentivize research into human-centered explanation methods, subsidize standards and certification infrastructure, and consider staged regulation balancing innovation with accountability in high-risk domains.
Prescriptive recommendations drawn by the paper's authors from the review of technical, social-science, and policy literatures; based on synthesis rather than empirical testing of policy impacts.
Clearer explanations and audit trails make it easier to assign responsibility and price risk (insurance markets, contract terms), potentially reducing uncertainty in public procurement and private contracts.
Economic and legal literature included in the review providing conceptual arguments and illustrative cases; no new empirical risk-pricing estimates provided in the paper.
Better explainability (when usable) raises willingness-to-adopt AI in regulated, risk-averse sectors by reducing information asymmetries and perceived liability—potentially expanding market size for explainable systems.
Economic and conceptual arguments synthesized from the reviewed literature; the review aggregates studies and arguments but does not present new quantitative adoption estimates.
Implementation requires organizational practices—governance, training, monitoring, and incentives—to translate explainability into safer, more legitimate AI use.
Synthesis of organizational, policy, and case-study literature in the review that identifies organizational measures correlated with effective deployment of explainable systems; descriptive evidence rather than causal experiments.
Regulatory frameworks, auditability, documentation (e.g., model cards, datasheets), and clear lines of responsibility amplify the effectiveness of explainability for accountability and compliance.
Synthesis of policy and governance literature included in the review that discusses how institutional mechanisms interact with technical explainability to produce accountability; descriptive evidence from case studies and governance proposals in the literature.
Labor demand will increasingly favor skills that support effective Human–AI teaming (interpretation, interrogation of AI, systems orchestration, shared-model building) rather than routine task execution.
Implication drawn from the framework and literature on complementarity and skill-biased technological change; presented as an expectation rather than quantified by labor market data in the paper.
Instituting continuous training, evaluation, and feedback loops is required to adapt Human–AI teams over time and maintain performance.
Prescriptive inference from organizational learning and human factors literature synthesized in the paper; suggested as best practice without empirical evaluation within the paper.
Building knowledge infrastructures that capture, curate, and make provenance accessible is necessary for team knowledge continuity, accountability, and learning.
Conceptual recommendation informed by literature on knowledge management and provenance; no empirical measures or case studies reported to quantify impact.
Partitioning roles — assigning pattern-detection tasks to AI and normative or contextual judgment to humans — improves task allocation based on comparative strengths.
Design recommendation derived from matching cognitive primitives to task types, supported conceptually by literature; not validated with empirical experiments in this paper.
Complementarity requires structuring interactions so humans and AI amplify each other's strengths rather than substitute for one another.
Conceptual argument based on theoretical review of complementarity and collective intelligence; no empirical tests included.
Aligning AI capabilities with human cognitive processes — reasoning, memory, and attention — is foundational to effective Human–AI teaming.
Theoretical grounding and literature synthesis drawing on cognitive science and human factors; proposed as a core lens for the framework rather than validated empirically in the paper.
Human–AI teams can achieve true complementarity such that joint team performance exceeds that of humans or AI alone.
Conceptual claim supported by an integrative, cross-disciplinary framework synthesizing literature from collective intelligence, cognitive science, AI, human factors, organizational behavior, and ethics. No primary empirical dataset or controlled experiments reported in the paper.
Operationalizing explainability alongside monitoring (data-drift detection, retraining schedules) and usage rules stabilizes managerial outcomes and raises adoption/trust.
Argument supported by the pilot illustration and the paper's operational design; evidence primarily from single-case pilot and conceptual reasoning rather than multi-site causal testing.
Explainability (XAI) tools were integrated with the model and, together with operational quality controls (data-drift monitoring, retraining routines, and usage regulations), increased user trust and improved reproducibility of managerial impact in the pilot.
Pilot case study reporting integration of XAI and operational controls and reporting increases in user trust and reproducibility of managerial outcomes (single SME pilot; qualitative and quantitative details referenced but not listed in the summary).
A pilot implementation in an SME for inventory-demand forecasting used a gradient-boosting model which outperformed a business-as-usual baseline on forecasting accuracy metrics.
Single pilot case study reported in the paper: inventory-demand forecasting pilot comparing a gradient-boosting model to a baseline forecasting approach (sample: one SME pilot; specific implementation details and exact metrics not provided in the summary).
Firms will reallocate investment toward cloud infrastructure, data engineering, model ops, and financial data integration, favoring vendors providing interoperable, audit-friendly solutions.
Predictive claim about investment incentives based on the paper's architectural and governance analysis; no spending data or vendor market-share evidence presented.
Next-generation financial analytics frameworks embed AI (ML, NLP, anomaly detection) into core financial systems to shift enterprises from retrospective reporting to predictive, prescriptive, and real-time decision-making.
This is the paper's central conceptual claim supported by a descriptive synthesis of AI techniques and system architecture; no empirical sample, controlled experiments, or deployment case data are presented—recommendations are justified by logical argument and examples of techniques.
Documented benefits of structured risk management include improved organizational resilience and stability under uncertainty.
Synthesis of claims in the literature reviewed; secondary cross-sectional evidence from peer-reviewed articles and practitioner sources within the ten-year scope (no primary quantitative validation in this review).
Transparent communication with stakeholders and the use of risk metrics/KPIs improve decision-making and stakeholder trust.
Thematic finding across reviewed articles and practitioner guidance; supported by references to reporting and KPI use in ISO/COSO-aligned literature.
Continuous monitoring and feedback loops enable learning and adaptation in risk management.
Identified as a recurring theme in the qualitative synthesis of the literature and embedded in recommended frameworks; based on secondary sources over the last ten years.
Use of formal frameworks and standards (ISO 31000, COSO ERM) helps ensure consistency and comparability in risk management practice.
Recommendation and frequent citation of formal frameworks in the reviewed literature and reference materials; thematic synthesis highlights frameworks as enablers of consistency.
Risk management functions as a strategic capability (not merely defensive), supporting sustainability and competitive advantage.
Recurring theme across the reviewed literature and alignment with established frameworks (ISO 31000, COSO ERM) identified via thematic analysis of the past ten years of publications and reference works.
Organizations that implement structured risk management processes experience greater stability, better decision-making, and higher stakeholder trust.
Qualitative literature review (thematic synthesis) of national and international journal articles, reference books, and risk frameworks (notably ISO 31000 and COSO ERM) from the past ten years; secondary cross-sectional literature evidence; no primary quantitative data or effect-size estimation reported.
AI reduces marginal labor needed for routine complaint handling, yielding cost savings and productivity gains, though savings depend on case mix and extent of automation.
Throughput metrics, reported reductions in manual processing from system logs, and administrator cost/performance reports; no standardized cost-effectiveness analysis provided across sites.
Hybrid models (AI-assisted triage + human adjudication for complex/sensitive cases) with governance, monitoring, and safeguards are the most sustainable approach.
Authors' best-practice recommendation synthesizing quantitative performance gains, qualitative stakeholder preferences, and observed challenges (privacy, bias, integration); supported by mixed-methods evidence but not tested as a randomized alternative.
Faster, clearer processes tend to raise patient satisfaction, particularly for routine queries.
Structured patient surveys measuring satisfaction and perceived clarity before/after AI adoption or between adopters/non-adopters; qualitative support from interview/open-ended survey responses (sample sizes/effect sizes not detailed).
System logs and dashboards improve transparency and managerial visibility into grievance workflows.
Platform logs and dashboard outputs analyzed for throughput and process-stage visibility; administrator interviews and surveys reporting improved oversight and traceability.
Automated classification increases consistency and accuracy of complaint categorization.
System-generated classification labels compared to human labels and/or prior categorizations using error rate/consistency metrics extracted from platform logs; supported by descriptive statistics (no specific effect sizes provided).
AI tools reduce complaint-response latency and speed up routing/triage.
Quantitative measurement from system logs and grievance records (timestamps for intake, triage, and response); analyses included before/after or adopter/non-adopter comparisons (exact sample size and statistical controls not reported here).