Evidence (2320 claims)
Adoption
5227 claims
Productivity
4503 claims
Governance
4100 claims
Human-AI Collaboration
3062 claims
Labor Markets
2480 claims
Innovation
2320 claims
Org Design
2305 claims
Skills & Training
1920 claims
Inequality
1311 claims
Evidence Matrix
Claim counts by outcome category and direction of finding.
| Outcome | Positive | Negative | Mixed | Null | Total |
|---|---|---|---|---|---|
| Other | 373 | 105 | 59 | 439 | 984 |
| Governance & Regulation | 366 | 172 | 115 | 55 | 718 |
| Research Productivity | 237 | 95 | 34 | 294 | 664 |
| Organizational Efficiency | 364 | 82 | 62 | 34 | 545 |
| Technology Adoption Rate | 293 | 118 | 66 | 30 | 511 |
| Firm Productivity | 274 | 33 | 68 | 10 | 390 |
| AI Safety & Ethics | 117 | 178 | 44 | 24 | 365 |
| Output Quality | 231 | 61 | 23 | 25 | 340 |
| Market Structure | 107 | 123 | 85 | 14 | 334 |
| Decision Quality | 158 | 68 | 33 | 17 | 279 |
| Fiscal & Macroeconomic | 75 | 52 | 32 | 21 | 187 |
| Employment Level | 70 | 32 | 74 | 8 | 186 |
| Skill Acquisition | 88 | 31 | 38 | 9 | 166 |
| Firm Revenue | 96 | 34 | 22 | — | 152 |
| Innovation Output | 105 | 12 | 21 | 11 | 150 |
| Consumer Welfare | 68 | 29 | 35 | 7 | 139 |
| Regulatory Compliance | 52 | 61 | 13 | 3 | 129 |
| Inequality Measures | 24 | 68 | 31 | 4 | 127 |
| Task Allocation | 71 | 10 | 29 | 6 | 116 |
| Worker Satisfaction | 46 | 38 | 12 | 9 | 105 |
| Error Rate | 42 | 47 | 6 | — | 95 |
| Training Effectiveness | 55 | 12 | 11 | 16 | 94 |
| Task Completion Time | 76 | 5 | 4 | 2 | 87 |
| Wages & Compensation | 46 | 13 | 19 | 5 | 83 |
| Team Performance | 44 | 9 | 15 | 7 | 76 |
| Hiring & Recruitment | 39 | 4 | 6 | 3 | 52 |
| Automation Exposure | 18 | 16 | 9 | 5 | 48 |
| Job Displacement | 5 | 29 | 12 | — | 46 |
| Social Protection | 19 | 8 | 6 | 1 | 34 |
| Developer Productivity | 27 | 2 | 3 | 1 | 33 |
| Worker Turnover | 10 | 12 | — | 3 | 25 |
| Creative Output | 15 | 5 | 3 | 1 | 24 |
| Skill Obsolescence | 3 | 18 | 2 | — | 23 |
| Labor Share of Income | 8 | 4 | 9 | — | 21 |
Innovation
Remove filter
We conducted a systematic review of 42 studies published between 2014 and 2025.
The paper's methods section reports selection and inclusion of 42 studies covering the period 2014–2025 (sample size = 42).
Digital–real integration and New Quality Productive Forces exhibit a significant bidirectional positive relationship (each variable positively and significantly promotes the other).
Empirical results from the GS3SLS spatial simultaneous equations model applied to the 30-province panel (2011–2022); paper reports statistically significant positive coefficients in both directions.
Entrepreneurs' expectations about future opportunities were significantly shaped by interpersonal influence (peer effects).
Quantitative analysis linking measures of interpersonal/peer exposure among entrepreneurs to reported expectations about future opportunities; analysis conducted within the >27,000 respondent sample across 43 countries.
Crisis adaptation among small business owners during COVID-19 was driven less by macroeconomic structure and more by social embedding (social networks, peer influence, and collective identities).
Comparative quantitative analysis of a survey sample of over 27,000 individual entrepreneurs in 43 countries using a novel socially embedded framework (networks, collective identities, normative motivations); empirical tests comparing explanatory power of social mechanisms versus macro-structural factors for adaptation outcomes.
Structural breaks in patenting dynamics are concentrated after 2010, consistent with an inflection in AI diffusion and commercialization.
Application of structural-break detection methods to patent filing time series (1980–2019) across domains; reported concentration of detected breakpoints after 2010. (Paper reports timing and clustering of breaks; exact statistical tests not enumerated in the summary.)
Patenting in AI-enhanced robotics experienced a sharp acceleration beginning in the early 2010s.
Observed marked upturn in the AI-enhanced robotics patent time series from the early 2010s onward (patent filings 1980–2019). Structural break tests applied to the time series identify an acceleration concentrated after 2010.
The paper issues a research agenda for economists: empirically develop instruments linking first‑person temporal reports with behavioral and neural proxies; theoretically incorporate subjective temporality into models of utility, human capital, attention economics, and platform competition; and evaluate policy accounting for temporal‑experience externalities.
Explicitly stated research agenda and methodological recommendations in the paper; no empirical follow‑up included.
Economists will need new empirical measures: validated instruments translating phenomenological constructs (e.g., Chronons) into observable proxies or composite indices for welfare and labor studies, facing standardization and comparability challenges.
Methodological recommendation and discussion in the paper; no empirical measure development or validation reported.
The paper proposes candidate mappings from subjective reports to neural/behavioral signatures (e.g., neural markers of attentional episodes, temporal binding windows) and suggests experimental paradigms to operationalize temporal units.
Methodological proposals and suggested experimental agendas in the paper; no implemented experiments or sample sizes reported.
The framework situates itself at the intersection of neurophenomenology, computational phenomenology, brain–computer interfaces, and human–AI teaming research.
Cross-disciplinary literature synthesis and conceptual mapping in the paper; descriptive claim with no empirical sampling (N/A).
The paper introduces symbolic operators—Chronons, Hexachronons, Metachronos—as theoretical units intended to bridge first-person phenomenology of temporal experience with third‑person neurotechnology descriptions.
Theoretical proposal and definitional introduction within the paper (conceptual development); no experimental validation or sample (N/A).
XChronos is a philosophical-epistemological framework arguing that transhumanism must place subjective temporality (lived time, presence, attention, meaning) at the center of design and evaluation.
Conceptual/philosophical analysis and literature synthesis presented in the paper; no empirical sample or dataset (N/A).
A Random Survival Forest built on curated cancer‑death‑related genes (CDRG‑RSF) achieved the best long‑term prognostic performance among 14 tested ML algorithms for pancreatic cancer, with 3‑ and 5‑year AUCs > 0.7.
Comparison of 14 ML survival algorithms on curated prognostic genes; Random Survival Forest (CDRG‑RSF) reported superior 3‑ and 5‑year AUCs exceeding 0.7 (exact sample sizes/cohort details not provided in summary).
Experimental knockdown of PSME3 reduced proliferation and invasion and increased apoptosis in LUAD cells, implicating the PI3K/AKT/Bcl‑2 pathway as a mediator.
Functional assays (gene knockdown experiments) reported in the PIGRS study showing decreased proliferation/invasion and increased apoptosis after PSME3 knockdown, with pathway analyses implicating PI3K/AKT/Bcl‑2.
Deep neural networks (DNNs) better captured cross‑study differential expression (DEA) signals when predicting miRNA from mRNA than sparse linear models (LASSO); for HIV the cross‑study log2 fold‑change (log2FC) correlation was approximately R ≈ 0.59 for the DNN approach.
Analysis on seven paired viral infection datasets (including WNV and HIV); compared DNNs vs. LASSO for mRNA→miRNA prediction; reported cross‑study log2FC correlation R ≈ 0.59 for HIV for the DNNs. Methods included differential expression signal recovery across studies.
An AI‑powered pipeline (EPheClass) produced a parsimonious saliva microbiome classifier for periodontal disease with AUC = 0.973 using 13 features.
EPheClass pipeline using ensemble ML (kNN, RF, SVM, XGBoost, MLP), centred log‑ratio (CLR) transform and Recursive Feature Elimination (RFE); reported performance AUC = 0.973 for periodontal disease model with 13 features (sample size not specified in summary).
Core supply‑chain management challenges targeted by simulation are production layout, product strategy, and managing volume and variety.
Survey and critique of simulation applications presented in the paper; conceptual taxonomy of application areas.
The paper proposes a 'manufacturing operation tree'—an organizationally structured framework—to guide development of more realistic, validated, and industry‑relevant simulation models.
Conceptual/modeling output in the paper (diagram and explanation of the manufacturing operation tree); theoretical development rather than empirical testing.
Standardizing datasets, benchmarks, and evaluation protocols (including real-time metrics and resource/latency measurements) is necessary to improve comparability and deployment relevance.
Surveyed inconsistencies and methodological shortcomings motivate the recommendation for standardization; many papers call for better benchmarks.
Hybrid architectures combining rule-based filters with ML classifiers and ensembles are used to improve detection performance and reduce false positives.
Comparative analysis and examples from the literature where multi-stage or hybrid pipelines are proposed and evaluated.
Econometric and causal-inference tools (difference-in-differences, instrumental variables, randomized encouragement designs) are needed to estimate long-term effects of personalized robot interventions.
Recommended methodological agenda for AI economists in the paper; no applied causal studies presented.
Research and deployment will require new datasets: longitudinal multimodal interaction logs, user preference surveys, simulated user populations, and ethically annotated datasets for fairness and safety evaluation.
Data & Methods recommendations based on identified empirical needs; no dataset release or analysis in this paper.
Measuring welfare impact of personalized robots requires going beyond engagement to include non-market outcomes such as well-being, autonomy, and mental health.
Methodological recommendation in the implications and evaluation sections; no empirical measures provided.
A/B testing and longitudinal field studies are necessary for real-world validation of robot personalization, and metrics should include welfare-oriented outcomes (well-being, trust) in addition to engagement.
Recommended evaluation strategy drawing from HRI and RS experimental standards; no field trials reported in this work.
Prior to live trials, offline RS evaluation metrics (precision/recall, NDCG), counterfactual/off-policy estimators, and simulated users should be used to validate personalization policies.
Methodological recommendation based on RS evaluation practices; no empirical comparison with live trials in robots presented.
Contextual bandits and counterfactual/off-policy learning can enable safe exploration and off-policy evaluation when adapting robot interactions from logged data.
Methodological synthesis referencing contextual bandit and counterfactual learning techniques from RS and causal inference; no robotic implementation experiments reported.
Sequence-aware recommenders (RNNs, Transformers, Markov/session-based models) are suitable for modeling session dynamics and short-term preference shifts in robot interactions.
Survey of sequence/temporal RS models and their typical use cases; conceptual recommendation only.
RS tooling covers long-term user profiles, short-term/session signals, context-awareness, multi-objective ranking, and evaluation methods suited for personalization at scale.
Review of recommender-systems methods and tooling in the literature; conceptual synthesis without empirical new data.
Recommender systems are specialized in representing, predicting, and ranking user preferences across time and contexts (e.g., collaborative filtering, content-based models, sequential/session models).
Established RS literature surveyed and cited as the basis for the claim; conceptual argument, no new experiments.
Breakthroughs in structure prediction arise from end‑to‑end deep models that combine evolutionary information (MSAs, coevolutionary signals), geometric constraints and equivariant architectures, and large‑scale pretraining on sequence databases.
Paper describes methodological components: end‑to‑end architectures using MSAs, SE(3)/E(3)-equivariant layers, transformer‑based pretraining on UniRef/UniProt/metagenomic catalogs; no quantitative ablation studies are provided in the text.
Algeria’s national approach centers on capacity building and technological independence as central security priorities in its AI strategy.
Analysis of Algeria’s national AI and security documents and related policy texts cited in the comparative case review.
The EU has developed a detailed, rights‑protective regulatory framework that includes procedural safeguards and explicit risk prohibitions for AI.
Qualitative document analysis of EU regulatory acts and strategies (e.g., bloc‑level AI regulatory proposals and legal texts) and comparative literature review.
Practical takeaway: economists should treat consent design as a lever that changes data availability and incorporate consent frictions into demand and production-side models; they should collaborate with HCI and legal scholars to design experiments capturing behavioral and welfare effects.
Recommendation from the workshop summary intended for economists; based on interdisciplinary discussions and agendas rather than tested interventions.
The workshop produced interdisciplinary outputs including personas, prototypes, and a research agenda to better align user capabilities and values with data-driven AI systems.
Documented workshop activities (Futures Design Toolkit, co-design, position papers) and stated expected deliverables in the workshop summary; these are reported outputs rather than evaluated outcomes.
The THETA project provides an interactive, reproducible analysis platform and open-source code (https://github.com/CodeSoul-co/THETA).
Explicit statement and URL in paper; code and platform availability claimed for reproducibility and interactive use.
THETA wraps modeling in an AI Scientist Agent framework (Data Steward, Modeling Analyst, Domain Expert) that simulates grounded-theory judgment and iterative refinement.
Detailed description of a three-role agent workflow in the methods section: Data Steward (ingestion/preprocessing), Modeling Analyst (modeling/hyperparameter tuning), Domain Expert (qualitative assessment/constant comparison).
THETA uses hybrid textual embeddings that combine pretrained foundation-model semantic structure with DAFT adaptations to better capture latent, domain-relevant meanings.
Method description of 'textual hybrid embeddings' combining base foundation encoders and DAFT-tuned parameters; asserted benefit for capturing latent domain meanings (no quantitative ablation reported in summary).
THETA adapts foundation embedding models to domain language using parameter-efficient LoRA fine-tuning (Domain-Adaptive Fine-Tuning, DAFT), avoiding full model retraining.
Method description: LoRA applied to foundation embedding models as the DAFT procedure; claim of parameter-efficient fine-tuning rather than end-to-end retraining (no compute benchmarks in summary).
Integrating AI (notably ML and NLP) meaningfully automates routine software engineering tasks across requirements management, code generation, testing, and maintenance.
Systematic literature review of prior AI-for-SE work combined with an empirical survey of software engineering professionals reporting usage and examples of tool-supported automation; sample size for the survey not specified in the summary.
An autoencoder-based ODE emulator that maps parameter values to latent trajectories can flexibly generate different solution paths conditioned on parameters.
Architecture and experiments: authors present a novel encoder/decoder ODE emulator that learns latent representation of trajectories and maps parameter vectors to latent trajectories; empirical examples provided (details not in summary).
A quantile emulator trained conditional on MCMC parameter draws can produce conditional quantile predictions without training a Bayesian neural network.
Method and empirical demonstration: paper describes and implements a quantile emulator (network trained to predict conditional quantiles across parameter draws).
The method is architecture-agnostic: uncertainty handling via parameter samples allows use of any deterministic neural-network architecture (e.g., quantile regressors, autoencoders) without specialized Bayesian layers.
Conceptual argument and demonstrations: authors implement a quantile emulator and an autoencoder-based ODE emulator as examples, showing the same uncertainty treatment applies to different network types.
By sampling training parameter vectors from a calibrated posterior (via MCMC), the surrogate avoids training on unphysical or implausible parameter configurations.
Design choice described in methods: MCMC sampling is used to draw parameter samples from the model-parameter distribution/posterior, thereby focusing training data on plausible regions; no experiments provided here quantify frequency of unphysical samples under alternative schemes.
Dataset and code (CFD, CFM, CFR) are publicly released.
Repository link provided in the summary (https://github.com/ZhengyaoFang/CFM) and paper states public release of dataset and code.
The Color Fidelity Dataset (CFD) is a large-scale dataset of over 1.3 million images containing both real photographs and synthetic T2I outputs, organized with ordered levels of color realism to support objective evaluation.
Dataset construction described in paper and repository: size stated as >1.3M images; contains a mixture of real photos and synthetic images annotated/organized with ordered realism labels enabling relative judgments of color fidelity.
The surrogate loop (build/update GP → select acquisition target → inner optimization → propose evaluation → evaluate with true model → update surrogate) can be parameterized so that inner objective and acquisition encode whether one seeks minima, saddles, or double-ended transitions.
Detailed methodological description in the paper of the six-step loop and how inner objectives/acquisition are changed to represent different search tasks; supported by example implementations in code.
The accompanying Rust code implements the same six-step surrogate loop across all applications, demonstrating practical reproducibility of the framework.
Authors state that pedagogical Rust code is provided showing the exact same loop running all applications; code repository accompanies the paper.
An adaptive trust radius constrains surrogate-guided steps to regions where the surrogate is reliable (trust-region control).
Methodological description of adaptive trust-radius control in the surrogate loop; used in experiments demonstrating improved reliability of steps proposed by the surrogate.
Acquisition criteria (active learning) drive which points are evaluated next; different acquisition functions implement the different search tasks (minimization, single-point saddles, double-ended searches).
Method section describing task-specific acquisition functions and their role in selecting evaluation points; implemented in the Rust code and used in experiments reported in the paper.
A unified Bayesian optimization framework—implemented as a six-step surrogate loop—handles minimization, single-point saddle searches, and double-ended saddle searches by changing only the inner optimization target and acquisition criterion.
Methodological description in the paper: presentation of a six-step surrogate loop (build/update GP → select acquisition target → inner optimization on surrogate → propose evaluation points → evaluate with true model → update surrogate) parameterized so inner objective and acquisition encode different tasks; accompanied by pedagogical Rust code implementing the same loop for all tasks.