Evidence (7560 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
9875 claims
Filter claims →
Productivity
8807 claims
Filter claims →
Governance
7870 claims
Filter claims →
Human-AI Collaboration
7560 claims
Filtered →
Org Design
4892 claims
Filter claims →
Innovation
4781 claims
Filter claims →
Labor Markets
4004 claims
Filter claims →
Skills & Training
3308 claims
Filter claims →
Inequality
2332 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 | 870 | 233 | 116 | 1066 | 2363 |
| Governance & Regulation | 976 | 451 | 218 | 133 | 1809 |
| Organizational Efficiency | 949 | 224 | 144 | 88 | 1416 |
| Technology Adoption Rate | 764 | 287 | 141 | 122 | 1325 |
| Research Productivity | 501 | 152 | 74 | 362 | 1101 |
| Output Quality | 542 | 216 | 69 | 69 | 896 |
| Decision Quality | 387 | 198 | 94 | 54 | 740 |
| Firm Productivity | 513 | 67 | 101 | 27 | 714 |
| AI Safety & Ethics | 249 | 303 | 73 | 36 | 667 |
| Market Structure | 190 | 192 | 134 | 27 | 548 |
| Task Allocation | 243 | 77 | 91 | 36 | 452 |
| Innovation Output | 291 | 33 | 55 | 20 | 401 |
| Skill Acquisition | 206 | 72 | 65 | 21 | 364 |
| Employment Level | 133 | 63 | 115 | 22 | 335 |
| Fiscal & Macroeconomic | 153 | 79 | 52 | 32 | 323 |
| Task Completion Time | 206 | 37 | 12 | 15 | 272 |
| Firm Revenue | 179 | 52 | 29 | 5 | 266 |
| Consumer Welfare | 130 | 76 | 47 | 13 | 266 |
| Inequality Measures | 48 | 137 | 51 | 6 | 242 |
| Worker Satisfaction | 101 | 81 | 25 | 13 | 220 |
| Error Rate | 84 | 110 | 11 | 5 | 210 |
| Wages & Compensation | 98 | 47 | 30 | 10 | 185 |
| Regulatory Compliance | 88 | 73 | 17 | 7 | 185 |
| Automation Exposure | 66 | 64 | 33 | 16 | 182 |
| Team Performance | 105 | 29 | 30 | 11 | 176 |
| Training Effectiveness | 109 | 22 | 14 | 21 | 168 |
| Developer Productivity | 114 | 21 | 14 | 8 | 158 |
| Job Displacement | 12 | 90 | 24 | 1 | 127 |
| Hiring & Recruitment | 57 | 9 | 9 | 5 | 80 |
| Skill Obsolescence | 6 | 56 | 9 | 1 | 72 |
| Social Protection | 43 | 17 | 8 | 2 | 70 |
| Creative Output | 35 | 21 | 9 | 4 | 70 |
| Labor Share of Income | 18 | 21 | 17 | 1 | 57 |
| Worker Turnover | 15 | 16 | — | 4 | 35 |
| Industry | — | — | — | 1 | 1 |
Human Ai Collab
Remove filter
We evaluate 36 models; the strongest, Claude Opus 4.7 under Claude Code, reaches only 45.9%.
Empirical evaluation reported by the authors: 36 models tested on JobBench; highest-performing model and its score (Claude Opus 4.7 under Claude Code achieves 45.9%).
Organizations implementing AI without responsible transition mechanisms may worsen workforce anxiety, skill obsolescence, inequality, and trust erosion.
Paper's theoretical/conceptual assertion about risks of poorly-managed AI adoption; no empirical validation reported in the excerpt.
The International Monetary Fund estimates that nearly 40% of global employment is susceptible to AI, with exposure rising to 60% in advanced economies owing to cognitive task-oriented jobs.
Cited IMF estimate reported in the paper (reference to an IMF analysis; no sample size given in the excerpt).
Tenure negatively relates to AI use (OR = 0.846 per category).
Reported odds ratio from logistic regression for tenure categories predicting AI use; OR = 0.846 per tenure category.
The requirement that review + expected rework attention be lower than manual completion attention is substantially more stringent than the requirement that AI merely generate faster drafts.
Comparative analytical argument based on the model's derived stability conditions (theoretical/model-based reasoning; no empirical sample reported).
Under congestion, reviewers rationally raise the risk threshold for checking AI outputs, reducing scrutiny precisely when it would matter the most.
Analytical implication derived from the queueing model presented in the paper (theoretical/model-based inference; no empirical validation reported).
Mean-based metrics (e.g., tasks completed per worker-hour or mean handle time) can misrepresent AI's effects in workflows where tasks accumulate and compete for scarce human attention.
Argument and analysis presented in the paper; theoretical reasoning and illustrative queueing model (no empirical sample reported).
Regardless of apparent performance advances in AI technology, human and environmental factors of the organization may substantially attenuate — or even negate — the effective productivity benefits.
Conceptual argument in the paper; theoretical reasoning and literature synthesis (no primary empirical data reported in the abstract).
Adopting AI in organizational practice does not guarantee productivity gains, because human and environmental factors critically moderate the relationship between AI deployment and realized productivity improvements.
Position paper's conceptual argument presented in the abstract; no empirical sample or quantitative study reported.
AI evaluation methods (benchmarks, red teaming, leaderboards) cannot be easily applied to human workers or yield comparable metrics.
Conceptual critique in the paper contrasting standard AI evaluation methods with human evaluation (no empirical comparisons provided).
Common criteria used to assess people (e.g., education, experience, references) cannot feasibly scale to AI systems.
Argumentative claim in the paper contrasting human hiring/evaluation practices with AI system assessment (conceptual; no empirical validation provided).
Human and machine workers may 'compete' for a given task, reproducing aspects of adversarial games.
Theoretical/assertional claim in the paper (conceptual discussion; no empirical data provided).
The increased use of algorithms in allocation decisions creates a Reverse Turing Test dynamic wherein the machine is now the judge.
Conceptual framing and argument presented in the paper (theoretical description; no empirical test reported).
A-insensitivity acts as a cognitive barrier between beliefs and trust (i.e., it reduces the extent to which beliefs about forecast accuracy are translated into trust).
Interpretation based on experimental findings showing that higher a-insensitivity weakens the predictive relationship between beliefs about accuracy and expressed trust in analysts (derived from measures and analyses in the lab experiment; sample size not reported in abstract).
Decision-makers who are more a-insensitive are less likely to incorporate their beliefs about forecast accuracy into their trust judgments.
Experimental data where participants' a-insensitivity was measured and used to predict the extent to which their beliefs (optimism about accuracy) translate into trust for analysts (moderation/interaction analysis implied; sample size not reported in abstract).
AI adoption presents workforce adaptation challenges.
Reported in the study's literature synthesis and thematic analysis of secondary sources (qualitative review). No sample size reported.
AI adoption raises ethical considerations.
Authors' thematic evaluation of secondary literature identifying ethical issues associated with human-AI collaboration (qualitative synthesis). No sample size reported.
AI adoption presents challenges related to skill gaps.
Thematic findings from peer-reviewed literature and secondary data (qualitative review). No sample size reported.
There is a 'speedup illusion' where people have accurate forecasts of independent completion times but significantly underestimate AI-assisted times.
Empirical pattern reported in the abstract: comparison of predicted vs. actual times shows accurate independent forecasts but underestimation of AI-assisted completion times (preregistered study, N = 1237).
The fidelity gain from richer profiles comes with more input tokens per call from the longer prompts they require (i.e., higher per-call input cost).
Measurement of input token counts per model call for prompt variants with and without life-history profiles in the benchmark experiments; comparison shows longer prompts require more input tokens.
A conventional two-arm test understates the algorithmic channel by a factor of two.
Empirical comparison reported in the paper between the three-arm design estimates and conventional two-arm test estimates from the live campaign.
In the same campaign, the creative channel moves female impression share by -0.68 ppt.
Empirical result from the live Meta campaign reported in the paper; measured effect size (-0.68 percentage points).
Adjusting for the realized audience is biased because audience is a post-treatment mediator.
Causal inference argument in paper explaining why conditioning on realized audience induces bias (audience as post-treatment mediator).
Every two-arm test conflates the creative's effect with the algorithm's targeting response.
Theoretical/causal argument presented in the paper about confounding in standard two-arm experiments when algorithmic delivery is endogenous.
As these systems scale, the bottleneck shifts away from raw model capability toward coordination.
Analytical/argumentative claim in the paper framing a shift in primary constraint; no empirical study or quantified benchmark reported.
More persuasive narratives may have had a detrimental effect on the ability to discriminate between a correct and incorrect AI prediction.
Exploratory analyses in the paper reporting reduced discrimination between correct and incorrect AI predictions when narratives were more persuasive.
More persuasive narratives may have had a detrimental effect on decision response times.
Exploratory analyses reported in the paper indicating persuasive narratives were associated with longer decision response times.
Higher benchmark performance does not reliably show that a system can carry out knowledge work in real-world deployment settings.
Argument based on review of current knowledge-work evaluation and benchmark design literature; paper motivates with conceptual analysis and references to empirical work showing mismatch between benchmark tasks and deployed work settings.
AI systems intended to simulate companionship or emotional responsiveness raise risks such as emotional manipulation, addictive interaction patterns, and potential impact of prolonged AI interaction on users’ mental well-being, particularly for vulnerable users.
Asserted risk statement in policy recommendations; no empirical study, prevalence data, or sample provided in the text.
Current systems still struggle with evidence preservation, reproducibility, weak-direction rejection, provenance tracking, cross-domain robustness, and accountable scientific closure.
Survey-identified recurring failure modes and limitations reported in literature and system descriptions; qualitative synthesis.
Current systems remain fragmented, differing in autonomy, domain scope, execution environment, validation mechanism, and human oversight.
Survey of existing systems and categorization across the listed dimensions; descriptive synthesis rather than an empirical meta-analysis.
Two of the top three leaderboard models (gpt-5 and claude) are noticeably more locally volatile than the third (gemini-3.1-pro), despite being close in overall strength.
Comparison of jaggedness/local volatility measures and overall scores from the tournament (top-three leaderboard).
Existing strategic-reasoning benchmarks evaluate models on fixed canonical games and may saturate as the frontier improves and fail to generalize to varied real-world strategic environments.
Conceptual critique stated in the paper's motivation/background; no empirical test reported in abstract.
Other changes are more nuanced and put the typical career growth opportunities, like receiving feedback from professional networks and promoting leadership and mentorship, at risk.
Qualitative reports from interview participants (n=24) expressing concerns that AI-driven changes may reduce feedback, leadership development, and mentoring opportunities.
Integrations of AI that neglect human factors are associated with increased anxiety, burnout, and disengagement among users.
Aggregate findings from the systematic review reporting associations in the literature between non-human-centered AI integration and negative psychological/work outcomes.
Notable challenges to AI implementation include concerns about algorithmic bias, privacy, transparency, job displacement, organizational culture, and issues related to ethical and legal oversight.
Synthesis of reported challenges across the 29 empirical studies included in the scoping review.
This transition proceeds without tools to forecast how individual employees will respond psychologically and behaviorally.
Asserted by the authors as a gap/need; no empirical inventory or systematic review presented in the excerpt to substantiate completeness of tool absence.
Workforce transformations are difficult to forecast and costly to mismanage.
Stated as a general assertion in the paper's introduction; no empirical data, sample, or formal analysis reported in the excerpt.
Student-designed tasks reveal hidden failures in current deep research systems: fluent, source-backed answers can still miss the right query, source, term, or evidence standard.
Qualitative analysis of failure modes from student-designed tasks and system evaluations reported in the paper (examples and discussion of how answers can be fluent and sourced yet incorrect on key criteria).
Evaluation on QuestBench shows that student-designed tasks reveal hidden failures in current deep research systems: across thirteen evaluated systems, the mean question-level pass rate is only 16.85%.
Empirical evaluation reported in the paper: 13 systems evaluated on QuestBench; aggregated mean question-level pass rate reported as 16.85%.
There is a negativity asymmetry: negative histories induce 1.62x more bias than positive (paired per item; t = 13.46, p < 10^-39, n = 2,481).
Paired per-item comparison of bias induced by negative versus positive histories; reported multiplicative factor, t-statistic, p-value, and sample size n = 2,481.
Static benchmarks capture only part of how large language models behave in practice.
Argument supported by the paper's experimental design comparing static evaluations with a timed multi-phase Risk environment that includes repeated planning/execution loops and real-system constraints.
The de-coring and skill-demand changes are concentrated among low entry-threshold, small firms.
Abstract statement reporting heterogeneity: concentration of observed patterns among firms characterized as small and with low entry thresholds.
Both displacement and augmentation exposure are associated with a de-coring pattern: a shallower and more dispersed skill portfolio with within-category importance diverging from share movements.
Empirical description in abstract that both forms of exposure correlate with changes in portfolio depth and dispersion, and with divergence between within-category importance and category shares.
Displacement exposure is negatively associated with the routine cognitive skill share.
Empirical result stated in abstract: negative association between displacement exposure and routine cognitive share, identified using within-firm variation and the constructed exposure measures.
In deployed settings, the effects of AI systems on human agency, creativity, and institutional well-being emerge over time, shaped by repeated interaction, reuse, and integration into real-world workflows, and these dynamics are rarely visible through pre-deployment evaluation or isolated prompt–response analysis.
Argumentative observation based on conceptual reasoning; no empirical data or sample size reported.
The most significant barriers to AI adoption reported by entrepreneurs are human-centred—talent scarcity, organisational resistance, and change management—rather than technology or cost alone.
Theme 'Barriers and the Adoption Journey' from thematic analysis of interviews (n=16); interviewees repeatedly cited human-centred barriers (talent scarcity, resistance, change management) over purely technical/cost barriers.
Raw interaction logs are inherently noisy, contain trial-and-error and low information density, and are inefficient for direct model training.
Author assertion describing properties of raw interaction logs; no empirical quantification provided in the excerpt.
Static 'human data' is expensive to scale and bounded by the knowledge of its creators.
Author claim/argument in the paper's introduction; no empirical sample or quantitative test reported in the provided text.
People exhibit self-estimate miscalibration: on average they believe they are using AI less than they actually are.
Same three pre-registered user studies (combined N = 2691) comparing participants' self-reported AI use against observed/recorded AI use during tasks.