Truace
Subscribe
The Index recomputed live from the record

What the evidence says.What the public feels.

The record holds 929 sourced gains and 770 sourced problems, averaging 68 and 66 on the index score. Readers have logged 22 public signals on the Pulse, which is kept apart and never counted as evidence.

929 gains
770 problems
Every sourced claim in the Index, one square each, shaded by the strength of its evidence. High Moderate Emerging

Ranks distinct AI gain and problem claims from the published record. Scores reward impact, independent source strength, scale, confidence, and recency.

1,699 results
Show filters and sorting

Download every matching row, not just this page:Export CSVExport JSON

AI gains · 929

37
HealthStableModerate evidence · 1 source

Meta-analysis of 19 studies with 100,790 participants found AI/ML models achieved pooled discrimination of 0.836 for predicting tuberculosis treatment failure.

By August 2026, a systematic review and meta-analysis of 34 studies evaluated AI and machine learning models to predict tuberculosis treatment failure. Nineteen studies with 100,790 participants were pooled, yielding an AUC of 0.836 with high heterogeneity, with tree-based and multimodal approaches common and most publications appearing after 2019.

Impact 30%
49
Evidence 25%
95
Scale 20%
85
Confidence 15%
87
Recency 10%
90

Updated Aug 18, 2026 · TRV-2026-0817

78Index score
38
HealthStableModerate evidence · 1 source

Patients recovering from orthopedic surgery reported 80.3% willingness to use patient-facing AI systems for transitional care, with priorities shifting to functional safety and rehabilitation guidance after discharge.

Researchers surveyed 752 orthopedic surgery patients across 33 hospitals in Guangdong, China, asking them to rate standardized descriptions of AI functions such as chatbots, vision-based monitoring, and wearables for education, motion correction, and risk alerts during the hospital-to-home transition. By August 2026 publication, 80.3% reported willingness to use such systems, with care priorities moving from information and instructions in hospital to functional safety and rehabilitation support at home.

Impact 30%
69
Evidence 25%
95
Scale 20%
60
Confidence 15%
87
Recency 10%
87

Updated Aug 5, 2026 · TRV-2026-0647

78Index score
39
PolicyStableModerate evidence · 1 source

In benchmark experiments reported by May 2026, the unified framework increased supply chain disruption-prediction accuracy to 94.1%, reduced demand-forecast error, and raised marketing campaign ROI from 14.2% to 45.3%.

By May 12 2026, researchers reported development and benchmark testing of a Unified Business Intelligence framework that combines machine learning, predictive analytics, and explainable AI for supply chain management, financial risk assessment, and digital marketing. The paper draws on 43 studies from 2023-2026 and reports measured improvements in disruption prediction, demand forecasting, credit-risk classification, and campaign targeting within experimental settings.

Impact 30%
69
Evidence 25%
95
Scale 20%
60
Confidence 15%
87
Recency 10%
83

Updated Jul 13, 2026 · TRV-2026-0134

78Index score
40
HealthStableModerate evidence · 1 source

Integration of multi-omics and machine learning can improve cardiovascular disease management by supporting definitive and early diagnosis, severity assessment, full-course risk stratification, and individualized prediction of drug and surgical benefit-risk to inform decisions.

Published August 13, 2026 as a peer-reviewed review in Frontiers in Cardiovascular Medicine, the article synthesizes recent advances using machine learning together with multi-omics to address cardiovascular disease heterogeneity where conventional one-size-fits-all strategies often yield limited benefit.

Impact 30%
49
Evidence 25%
95
Scale 20%
85
Confidence 15%
87
Recency 10%
89

Updated Aug 16, 2026 · TRV-2026-0784

77Index score

AI problems · 770

37
HealthStableModerate evidence · 1 source

Most classical AI models for bipolar disorder treatment optimization had high risk of bias and lack of external validation, and remain exploratory rather than ready for clinical use.

A PRISMA-guided systematic review of 35 studies examined classical AI for treatment optimization in adult bipolar disorder across five outcomes: acute response, long-term maintenance, relapse/readmission, safety/dose, and brain aging/phenotyping. By the July 2026 publication date, pooled performance ranged from modest for acute response (AUC 0.68) to moderate-to-high for maintenance (AUC 0.80) and high accuracy for safety/dose (85%-97%).

Impact 30%
49
Evidence 25%
95
Scale 20%
85
Confidence 15%
87
Recency 10%
84

Updated Jul 22, 2026 · TRV-2026-0511

77Index score
38
HealthStableModerate evidence · 1 source

Deployment of AI in healthcare is limited by risks of data privacy breaches, algorithmic bias, lack of model interpretability, and gaps in regulatory oversight and human clinical oversight.

Published September 23 2025 as a peer-reviewed review, the article surveys how AI is being applied across healthcare, from analyzing electronic health records and medical imaging to supporting drug discovery, predictive analytics, telemedicine and wearable biosensors, with emphasis on low-resource and remote settings.

Impact 30%
49
Evidence 25%
95
Scale 20%
85
Confidence 15%
87
Recency 10%
84

Updated Jul 22, 2026 · TRV-2026-0484

77Index score
39
CrimeStableModerate evidence · 1 source

Same machine learning fraud detection systems encounter persistent challenges including data imbalance, concept drift and privacy concerns that complicate implementation in operational financial environments.

On 2025-11-05, Applied Sciences published a comprehensive review of machine learning for financial fraud detection. The authors surveyed supervised, unsupervised and hybrid approaches across credit card, financial statement, insurance and money laundering fraud, reviewed datasets and metrics, and included two case studies applying supervised models to real-world banking data.

Impact 30%
49
Evidence 25%
95
Scale 20%
85
Confidence 15%
87
Recency 10%
84

Updated Jul 22, 2026 · TRV-2026-0477

77Index score
40
ClimateStableModerate evidence · 1 source

Large-scale deployment of AI servers across the United States is projected to create 731 to 1,125 million m3 of annual water use and 24 to 44 Mt CO2-equivalent of additional annual carbon emissions between 2024 and 2030, jeopardizing net-zero goals.

Published November 10 2025 in Nature Sustainability, the study models the sustainability implications of rapidly expanding generative AI server installations across the United States, projecting annual water and carbon footprints through 2030 and testing mitigation options.

Impact 30%
49
Evidence 25%
95
Scale 20%
85
Confidence 15%
87
Recency 10%
84

Updated Jul 22, 2026 · TRV-2026-0474

77Index score

Recomputed live from the record · Oct 11, 2026, 4:15 AM