The Commonplace
Home Dashboard Papers Evidence Syntheses Digests 🎲

Evidence (7953 claims)

Adoption
5539 claims
Productivity
4793 claims
Governance
4333 claims
Human-AI Collaboration
3326 claims
Labor Markets
2657 claims
Innovation
2510 claims
Org Design
2469 claims
Skills & Training
2017 claims
Inequality
1378 claims

Evidence Matrix

Claim counts by outcome category and direction of finding.

Outcome Positive Negative Mixed Null Total
Other 402 112 67 480 1076
Governance & Regulation 402 192 122 62 790
Research Productivity 249 98 34 311 697
Organizational Efficiency 395 95 70 40 603
Technology Adoption Rate 321 126 73 39 564
Firm Productivity 306 39 70 12 432
Output Quality 256 66 25 28 375
AI Safety & Ethics 116 177 44 24 363
Market Structure 107 128 85 14 339
Decision Quality 177 76 38 20 315
Fiscal & Macroeconomic 89 58 33 22 209
Employment Level 77 34 80 9 202
Skill Acquisition 92 33 40 9 174
Innovation Output 120 12 23 12 168
Firm Revenue 98 34 22 154
Consumer Welfare 73 31 37 7 148
Task Allocation 84 16 33 7 140
Inequality Measures 25 77 32 5 139
Regulatory Compliance 54 63 13 3 133
Error Rate 44 51 6 101
Task Completion Time 88 5 4 3 100
Training Effectiveness 58 12 12 16 99
Worker Satisfaction 47 32 11 7 97
Wages & Compensation 53 15 20 5 93
Team Performance 47 12 15 7 82
Automation Exposure 24 22 9 6 62
Job Displacement 6 38 13 57
Hiring & Recruitment 41 4 6 3 54
Developer Productivity 34 4 3 1 42
Social Protection 22 10 6 2 40
Creative Output 16 7 5 1 29
Labor Share of Income 12 5 9 26
Skill Obsolescence 3 20 2 25
Worker Turnover 10 12 3 25
ML models can continuously derive available gigs and demand signals from marketplace activity, producing up-to-date opportunity lists and predicted wages.
Implemented ML models ingest real-time market activity/platform signals in the pilot to generate opportunity lists and wage predictions; no reported out-of-sample accuracy or prediction error metrics in the summary.
medium positive AI-Driven Skill Mapping and Gig Economy Matching Algorithm f... availability/recency of opportunity lists; accuracy of predicted wages
Skills can be inferred from multiple nontraditional inputs—self-reported information, short-term work histories, and community recommendations—creating richer profiles beyond formal work experience.
System design uses NLP to normalize and extract skills from profiles, short-term work records, and community recommendations; claim is supported by the implemented data integration approach rather than by quantified external validation in the summary.
medium positive AI-Driven Skill Mapping and Gig Economy Matching Algorithm f... inferred skill coverage/quality or profile richness
The pilot implementation produced higher reported wages for youth matched through the system relative to baseline informal methods.
Pilot comparison reported higher reported wages for matched youth; summary lacks sample size, measurement protocol, and statistical inference.
medium positive AI-Driven Skill Mapping and Gig Economy Matching Algorithm f... reported wages (self-reported earnings)
The pilot implementation led to higher correct matches compared to existing informal search methods.
Pilot deployment compared matching accuracy versus baseline informal job-search approaches; the paper summary reports a 'marked increase' but provides no numerical details, sample size, or significance levels.
medium positive AI-Driven Skill Mapping and Gig Economy Matching Algorithm f... matching accuracy / proportion of correct matches
AI-driven NLP and ML can substantially reduce search frictions in Nairobi’s informal and gig economies by dynamically deriving individual skills and real-time market opportunities, then algorithmically matching youth to short-term work.
Pilot implementation of an end-to-end system combining NLP, ML and a matching algorithm deployed in Nairobi and compared qualitatively/aggregately against baseline informal search methods; paper summary does not report sample size, statistical tests, or numerical effect sizes.
medium positive AI-Driven Skill Mapping and Gig Economy Matching Algorithm f... search frictions (reduction), matching quality
Firms should pair strong-performing ensemble/deep models with explainability tools (e.g., feature-importance, SHAP) and fairness audits, and prefer pilot human-in-the-loop implementations to validate economic impacts and reduce operational risks.
Authors' practical recommendations based on empirical model performance, interpretability analyses, and noted limitations; presented as guidance rather than empirically validated interventions.
medium positive Adoption of AI-Based HR Analytics and Its Impact on Firm Pro... Recommended practices for deployment (procedural guidance, not an outcome metric...
Variable-contribution analyses (feature importance / model explanation techniques) clarified which inputs drive predictions, making results actionable for HR decision-making.
The paper reports use of feature-importance and model-explanation methods to quantify variable contributions and interpretable outputs intended for HR practitioners.
medium positive Adoption of AI-Based HR Analytics and Its Impact on Firm Pro... Interpretability outputs (feature importance / explanation scores) linked to job...
Employee engagement/participation levels, learning agility (pace of acquiring new skills), tenure in current role, and perceived workload/manageability are consistently among the most important predictors of job performance in the datasets examined.
Feature-importance and model-explanation analyses (e.g., feature importance, SHAP-style approaches) applied across multiple publicly available workforce datasets produced consistently high importance scores for these variables.
medium positive Adoption of AI-Based HR Analytics and Its Impact on Firm Pro... Variable importance for predicting job performance
The models' superior performance hinges on their ability to capture complex, non-linear patterns in features (e.g., engagement, learning agility, tenure, workload perception).
Inference from comparative model performance: non-linear models (ensembles, DNNs) outperform linear baselines; feature engineering captured engagement dynamics and learning trends; variable-contribution analyses highlighted these feature types as influential.
medium positive Adoption of AI-Based HR Analytics and Its Impact on Firm Pro... Contribution of non-linear feature interactions to predictive performance (refle...
These predictive gains persist when models are applied to different company datasets, indicating better generalization of AI methods.
Cross-company tests described in the paper: models trained/tuned on one dataset and evaluated on others (holdout across organizations) with reported performance metrics demonstrating persistent improvements for AI methods.
medium positive Adoption of AI-Based HR Analytics and Its Impact on Firm Pro... Out-of-sample predictive performance across datasets/companies (AUC, F1, accurac...
Responsible implementation requires legal/liability clarity, continuous monitoring for performance drift and distributional shifts, usable explanations, baseline AI literacy for clinicians, and co-design with frontline radiology teams.
Synthesis of governance literature, implementation best-practice reports, and recommendations from usability and deployment studies.
medium positive Human-AI interaction and collaboration in radiology: from co... successful deployment metrics, monitoring alerts for drift, clinician comprehens...
Triage and automation can shorten time-to-diagnosis, increase throughput, and reduce time spent on repetitive tasks.
Observational deployment reports and simulation studies that measured time-to-report or throughput improvements in pilot settings (evidence heterogeneous and context-dependent).
medium positive Human-AI interaction and collaboration in radiology: from co... time-to-diagnosis, studies-per-hour per radiologist, time spent on repetitive ta...
Integration points for AI across the imaging pathway include acquisition (image quality/protocol selection), triage (prioritization), interpretation/reporting (detection, quantification, report pre-population), and post-interpretation (teaching, QA, model improvement loops).
Descriptive synthesis of reported implementations and proposed use cases in the literature and deployment reports across multiple institutions.
medium positive Human-AI interaction and collaboration in radiology: from co... site-level implementation metrics by workflow stage (e.g., reduced repeat scans,...
Human-AI collaboration can produce synergistic gains (diagnostic complementarity) when errors are uncorrelated and tasks are allocated to leverage comparative strengths.
Theoretical/analytical models of error complementarity and empirical reader studies showing instances where combined readings outperform either agent alone (evidence drawn from multiple small-to-moderate reader studies and simulations).
medium positive Human-AI interaction and collaboration in radiology: from co... combined diagnostic accuracy (aggregate sensitivity/specificity), reduction in m...
AI in radiology has clear potential to improve diagnostic performance and workflow efficiency.
Narrative synthesis of laboratory evaluation studies, reader/comparison studies, and a limited number of observational deployment reports showing improved algorithm accuracy and some improvements in measured throughput or time-to-review in pilots (study sizes and settings heterogeneous; few large-scale RCTs).
medium positive Human-AI interaction and collaboration in radiology: from co... diagnostic accuracy (sensitivity/specificity), workflow efficiency (throughput, ...
Cognitive Shadow supports real-time model updates based on immediate user feedback, enabling iterative improvement and continuous alignment with human decision patterns.
Described human-in-the-loop interaction loop where CS captures human decisions, provides recommendations, receives immediate feedback, and updates models dynamically in the simulation environment (implementation detail).
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... model update frequency / change in model-human agreement over iterative interact...
HACL/CS reduces omission rates (missed detections) in the simulated scenarios.
Omission/error rates were tracked and compared between conditions in the simulated testbed; summary claims reduction in omissions with HACL assistance but does not report numeric effect sizes or significance.
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... omission rate / missed detections
HACL/CS reduces time-to-decision in the simulated maritime surveillance tasks.
Measured time-to-classify in simulation under human-alone vs HACL-assisted conditions; summary indicates reductions in time-to-decision but lacks detailed statistics in the provided description.
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... time to classify / time-to-decision
In the simulated Canadian Arctic maritime surveillance domain, HACL/CS shows promise for improving classification accuracy.
Performance comparison between human-alone and HACL-assisted conditions in the maritime surveillance simulation measuring classification accuracy; summary reports improvement but does not provide sample size or significance levels.
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... classification accuracy (correctly classifying entities/threat levels)
Adjustable autonomy via self-confidence thresholds enables the system to act autonomously on high-certainty predictions and defer to humans on low-certainty cases.
System design feature of Cognitive Shadow implemented in simulation: autonomy decision rule based on meta-model confidence thresholds; behavior demonstrated in human-in-the-loop scenarios.
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... frequency of autonomous actions vs. human deferrals as a function of meta-model ...
The Cognitive Shadow toolkit quantifies AI reliability with an empirical (0–1) confidence metric produced by a recursive meta-model.
Design and implementation detail: primary supervised models are paired with a recursive meta-model that predicts the primary model's reliability per situation and outputs a 0–1 empirical confidence score; applied in the simulated testbed.
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... meta-model predicted reliability (empirical confidence score, 0–1)
Implementing an adaptive command-and-control process augmented by AI metacognition (the Cognitive Shadow toolkit) aligns AI judgments with expert human decision patterns.
Cognitive Shadow (CS) implemented as supervised ML models trained to mimic expert human decisions in the simulated maritime scenarios; alignment assessed by comparing model outputs to human expert decisions during human-in-the-loop interaction (implementation validated in simulation).
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... degree of alignment between AI model judgments and expert human decision pattern...
Human-AI co-learning (HACL) improves human-autonomy teaming (HAT) effectiveness.
Evaluated in a simulated Canadian Arctic maritime surveillance testbed using human-in-the-loop experiments comparing human-alone vs HACL-assisted conditions; exact participant sample size and statistical details not provided in the summary.
medium positive Human Autonomy Teaming and AI Metacognition in Maritime Thre... overall HAT effectiveness (operational performance and human factors composite)
Self-directed autonomous agents (those that autonomously generated prompts and selected tools) bypassed human prompting failures and outperformed most human teams on the challenge set.
Comparative analysis of the four autonomous agents' trajectories, tool use, and success rates versus the 41 human participants/teams on the same fresh challenges; observed correlation between autonomous self-direction and higher success relative to most teams.
medium positive Understanding Human-AI Collaboration in Cybersecurity Compet... challenge solving rates and relative rankings of self-directed agents versus hum...
Clinical-interface validation with real physicians on mobile devices confirmed the practical viability and usability of the HADT system and interface.
Paper reports an interface test with real doctors using the mobile interface as part of validation (number of physicians, tasks, and quantitative usability metrics not specified in the summary).
medium positive Hierarchical Reinforcement Learning Based Human-AI Online Di... practical viability / usability in clinical-interface testing (physician interac...
On public datasets HADT achieves superior accuracy/human-effort trade-offs compared to baselines (fully human, fully automated, and simpler assignment strategies).
Comparative evaluations reported in the paper on public medical-consultation datasets (baselines listed broadly; exact baseline implementations, dataset names, and quantitative comparisons not included in the provided summary).
medium positive Hierarchical Reinforcement Learning Based Human-AI Online Di... diagnostic accuracy and human effort relative to baseline methods
The execution machine uses masked hierarchical reinforcement learning with bottom-up training to ask informative symptom questions from a large symptom space.
Methodological description and reported training procedure in the paper (bottom-up training applied to the execution module; claimed to improve question selection; experimental validation referenced but details not provided in the summary).
medium positive Hierarchical Reinforcement Learning Based Human-AI Online Di... quality/informativeness of symptom questions; downstream diagnostic accuracy
A two-layer hierarchical reinforcement learning system—an assignment 'master' and an execution 'machine' (plus human doctors)—effectively balances accuracy and human cost.
Architectural design and experimental evaluation in the paper demonstrating trade-offs between diagnostic accuracy and human involvement using the hierarchical RL setup (experiments run on public datasets; exact sample sizes not given).
medium positive Hierarchical Reinforcement Learning Based Human-AI Online Di... trade-off between diagnostic accuracy and human effort
The Human-AI Diagnostic Team (HADT) framework can deliver near-expert-level online symptom inquiry and diagnosis while using very little human labor.
Performance reported on public datasets and clinical-interface tests with real physicians; described comparisons to expert-level performance in the paper's main finding (specific comparative statistics to experts not provided in the summary).
medium positive Hierarchical Reinforcement Learning Based Human-AI Online Di... quality of symptom inquiry / diagnostic performance (compared to expert-level)
HADT reached up to 89.4% diagnostic accuracy while requiring only 10.9% human effort.
Empirical results reported in the paper from experiments on public online medical-consultation datasets and clinical-interface validation with real doctors (dataset names and sample sizes not specified in the provided summary).
medium positive Hierarchical Reinforcement Learning Based Human-AI Online Di... diagnostic accuracy; human effort (proportion of turns/time requiring human doct...
HR and organizational implication: Firms should consider emotional dynamics when designing hybrid teams; training, monitoring, and pairing strategies (human–human, human–AI) matter for short-term task performance.
Interpretation by authors based on experimental findings that emotion and perceived service empathy alter collaboration proficiency across partner types in temporary virtual tasks (n = 861).
medium positive Adoption of AI partners in temporary tasks: exploring the ef... short-term task performance / collaboration proficiency
Design implication: Investing in AI features that convey empathy or supportive social cues could increase collaboration proficiency when emotion matters.
Authors' inference grounded in mediation (service empathy → collaboration proficiency) and the partner-type moderation of that effect in the experiment (n = 861), suggesting empathy-signaling features could alter outcomes.
medium positive Adoption of AI partners in temporary tasks: exploring the ef... collaboration proficiency
Findings are consistent with the 'computers are social actors' (CASA) framework: people respond to computers as social actors, so social/affective cues (not just whether a partner is human) shape collaboration outcomes.
Theoretical interpretation offered by the authors, supported by empirical patterns in the experiment (significant moderation by emotion and mediation by service empathy despite no main effect of partner type).
medium positive Adoption of AI partners in temporary tasks: exploring the ef... collaboration outcomes (collaboration proficiency)
Policy/managerial implication: organizational structures and incentives (e.g., TMT diversity, ESOPs) are effective levers to sustain managerial attention to employee welfare and mitigate the negative effects of deep AI penetration on ECSR.
Inference from empirical moderator results (TMT diversity and ESOP interactions) combined with theoretical ABV/dual-agent argument; paper includes managerial and policy recommendations based on these findings.
medium positive Attention to Whom? AI Adoption and Corporate Social Responsi... ECSR (and managerial attention as targeted by interventions)
Employee stock ownership (ESOP) moderates the relationship by flattening and right-shifting the inverted U, aligning employee incentives and preserving employee-focused attention as AI adoption deepens.
Interaction terms between AI (and AI^2) and ESOP presence/level show mitigated negative effects of high AI adoption on ECSR and a later turning point; based on panel regressions with controls and robustness checks on the 2,575-firm sample.
Top management team (TMT) functional diversity moderates the AI–ECSR curve by flattening it and right-shifting the peak, delaying and mitigating negative attention shifts from employees to AI.
Interaction of AI (and AI^2) with a TMT functional diversity measure in panel regressions indicates a less pronounced inverted U and a later turning point for firms with more diverse TMTs; analysis uses the main panel (2,575 firms, 2013–2023) with robustness checks.
At low-to-moderate levels of AI adoption, AI increases managerial attention to employees and raises ECSR (human attention gain mechanism).
Positive slope of the estimated AI–ECSR relationship at lower AI values implied by the significant linear AI term in the quadratic panel model; theoretical framing via an attention-based view (ABV) and dual-agent model; empirical results interpreted as consistent with increased managerial attention and higher ECSR at low-to-moderate AI adoption. (Sample: 2,575 firms, 2013–2023.)
medium positive Attention to Whom? AI Adoption and Corporate Social Responsi... ECSR (and managerial attention as a theoretical/mediating construct; managerial ...
AI-mediated collaboration will create new organizational roles and governance structures, such as AI mediators and verification/oversight roles.
Conceptual discussion of organizational implications and illustrative role examples; no organizational case studies with sample sizes reported.
medium positive AI as a universal collaboration layer: Eliminating language ... emergence of new roles (count/frequency) and governance structures within organi...
Autonomous AI agents can automate routine coordination tasks, follow-up, and some task execution, thereby reducing human coordination overhead.
Paper uses conceptual mapping of agent capabilities to coordination/execution functions and provides illustrative case scenarios; no experimental or field data presented.
medium positive AI as a universal collaboration layer: Eliminating language ... human coordination time / routine task overhead / automated task completion rate
Multimodal systems (integrating text, speech, images, video) broaden communication channels and thus can improve the range and fidelity of mediated communication.
Conceptual argument and illustrative examples in the paper describing how multimodal integration maps to communication functions; no empirical validation reported.
medium positive AI as a universal collaboration layer: Eliminating language ... breadth/fidelity of communication channels; information transmission quality
Multilingual language models reduce language barriers by translating and normalizing meaning across languages.
Conceptual synthesis of capabilities (multilingual LMs) and mapping to coordination function (translation/normalization); supported in paper by illustrative examples rather than empirical testing.
medium positive AI as a universal collaboration layer: Eliminating language ... degree of language barrier reduction / fidelity of cross-language meaning transf...
Trust in AI should be conceptualized as a socio-technical, team-level mechanism (trust calibration) that mediates between AI design/enablers and downstream collaboration and performance, rather than an individual-level stable attitude.
Theoretical synthesis combining findings from the thematic analysis of 40 interviews with socio-technical systems theory (STS) and adaptive structuration theory (AST) to propose an initial and revised conceptual model linking enablers → trust-calibration practices → collaboration dynamics → performance.
medium positive AI in project teams: how trust calibration reconfigures team... conceptual framing (mediating mechanism linking design/enablers to collaboration...
Five enablers support effective trust calibration: transparency/explainability, clear role definitions, good user experience (UX), supportive cultural norms, and timely system feedback.
Synthesized from recurring themes in the interview data (N=40) where respondents identified these factors as facilitating appropriate reliance on AI in project settings; coded and aggregated through thematic analysis.
medium positive AI in project teams: how trust calibration reconfigures team... quality/appropriateness of trust calibration
Performance and reward structures must be redesigned to value oversight, hypothesis testing, escalation and governance behaviours that mitigate model risk but may not immediately increase output.
Managerial recommendation derived from the framework and organizational reward literature; no empirical evaluation provided.
medium positive Symbiarchic leadership: leading integrated human and AI cybe... alignment of incentives; frequency of oversight/governance behaviours; mitigatio...
Firms need new metrics to decompose value created by humans, AI, and their interaction (to distinguish complementarities versus substitution).
Analytic implication derived from the framework and literature on productivity measurement; presented as a recommendation for empirical work rather than tested evidence.
medium positive Symbiarchic leadership: leading integrated human and AI cybe... accuracy of productivity attribution; measurement of human–AI complementarities/...
Symbiarchic leadership is a practical, HR‑oriented framework for leading integrated human–AI “cyber teams,” specifying four linked leadership practices that make AI a co‑actor in knowledge work while preserving human judgement, accountability and organizational legitimacy.
Paper's central proposition based on theoretical synthesis of academic literature on human–AI collaboration, hybrid teams and digital‑era leadership plus illustrative practitioner examples; no original empirical data or experiments.
medium positive Symbiarchic leadership: leading integrated human and AI cybe... ability to lead integrated human–AI teams; preservation of human judgement, acco...
Policies improving data sharing, standardization, and model transparency would increase overall welfare by reducing duplication and improving model performance.
Policy argumentation in the paper drawing on economic theory and examples where shared datasets/standards improved research productivity.
medium positive Has AI Reshaped Drug Discovery, or Is There Still a Long Way... research productivity and welfare as affected by data-sharing, standardization, ...
Organizations that tightly integrate AI teams with experimental groups achieve higher productivity.
Case studies and internal metrics cited in the paper showing improved throughput and candidate progression in integrated teams versus siloed approaches.
medium positive Has AI Reshaped Drug Discovery, or Is There Still a Long Way... organizational productivity (throughput, candidate progression) as a function of...
Value accrues to firms that control high-quality data, integrated platforms, and wet-lab validation—data and experimental capacity are strategic assets.
Market and organizational analysis in the paper citing examples of firms leveraging proprietary data/platforms and wet-lab capabilities to advance candidates more effectively.
medium positive Has AI Reshaped Drug Discovery, or Is There Still a Long Way... firm success/value correlated with possession of high-quality data, integrated p...
AI reduces time and cost in early-stage discovery (discovery-to-candidate), lowering per-candidate screening and design costs.
Reported case studies and cost/time comparisons in the paper showing faster candidate identification and reduced experimental burden in early stages; aggregated industry claims summarized.
medium positive Has AI Reshaped Drug Discovery, or Is There Still a Long Way... time and monetary cost from discovery to candidate selection; per-candidate scre...