Truace
Subscribe
The Index recomputed live from the record

What the evidence says.What the public feels.

The record holds 933 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.

933 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.

1,703 results
Show filters and sorting

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

AI gains · 933

221
EducationStableHigh evidence · 3 sources

AI methods enable personalized sustainability courses and tools like adaptive learning systems and carbon footprint calculators for environmental literacy.

Published February 19, 2026, this PRISMA-based literature review examined how artificial intelligence intersects with environmental education to advance Quality Education and Climate Action goals. It surveyed applications of machine learning, natural language processing, computer vision and intelligent tutoring systems across different educational settings.

Impact 30%
49
Evidence 25%
100
Scale 20%
35
Confidence 15%
100
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0368

70Index score
222
CrimeStableHigh evidence · 5 sources

Literature shows machine learning models are applied to detect financial fraud, with credit card fraud detection models most widely used.

Published September 3, 2024, this peer-reviewed literature review applied PRISMA and Kitchenham methods to 104 articles from 2012-2023 obtained from Scopus, IEEE Xplore, Taylor & Francis, SAGE, and ScienceDirect to map how machine learning is used to detect financial fraud and its types.

Impact 30%
49
Evidence 25%
100
Scale 20%
35
Confidence 15%
100
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0366

70Index score
223
ClimateRisingHigh evidence · 5 sources

Artificial intelligence applied to planning and operation of distributed energy systems enhances the efficiency and reliability of electrical grids.

This 2024 peer-reviewed review paper surveys how artificial intelligence techniques including machine learning, optimization, and cognitive computing are being applied to the planning and operation of distributed energy systems in smart grids, covering prediction, optimization, resource coordination, renewable integration, and demand response.

Impact 30%
49
Evidence 25%
100
Scale 20%
35
Confidence 15%
100
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0361

70Index score
224
ScienceStableHigh evidence · 5 sources

Researchers developed a checklist for comprehensive assessment of ethical discussions in GenAI research that can be integrated into peer review to enhance research and support ethics disclosures for health-care applications.

Published September 17, 2024, this scoping review in The Lancet Digital Health examined ethical discussions surrounding generative AI in health care, including ChatGPT and other models used to synthesise data such as images for research and practical purposes. The authors found that ethical concerns have been widely noted but not translated into operational solutions.

Impact 30%
49
Evidence 25%
100
Scale 20%
35
Confidence 15%
100
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0360

70Index score

AI problems · 770

221
HealthStableModerate evidence · 1 source

AI model performance dropped in external validation to AUC 0.55-0.72, and multimodal integration did not show translated incremental benefit in TEST and EXVAL sets.

The I3LUNG study enrolled 2,396 patients with non-small cell lung cancer to develop AI models for immunotherapy selection, integrating clinical and blood data, CT, digital pathology and genomics into early and intermediate fusion models. CB-only models reached AUC up to 0.77 in the independent TEST set and outperformed PD-L1, ECOG PS, NLR, LDH and LIPI, and a usability study found physicians improved predictions with the explainable AI tool.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
95

Updated Sep 15, 2026 · TRV-2026-1095

68Index score
222
HealthStableModerate evidence · 1 source

Laboratories adopting AI face distinctive challenges including preanalytical variability, interplatform calibration differences, specimen quality effects, reagent-lot sensitivity, and context-dependent result interpretation that complicate validation and deployment.

This narrative review from Chinese Medical Journal examines how AI in laboratory medicine has evolved from conventional machine learning on structured results to deep learning and large language models that handle unstructured clinical text. It compares four paradigms and reviews evidence across blood cell morphology, autoverification, infectious risk stratification, urinalysis interpretation, decision support, and report generation.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
95

Updated Sep 15, 2026 · TRV-2026-1093

68Index score
223
HealthStableModerate evidence · 1 source

Large language models could distribute misinformation and exacerbate scientific misconduct in medicine due to lack of accountability and transparency.

A Communications Medicine overview published October 10 2023 examines large language models as text-processing AI tools that gained wide attention after ChatGPT's November 2022 release, assessing their near-human ability to answer, summarize and translate and their emerging use in clinical practice, medical research and medical education.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
95

Updated Sep 14, 2026 · TRV-2026-1089

68Index score
224
HealthStableModerate evidence · 1 source

Some algorithms overestimated the number at risk versus FRS without addressing overdiagnosis risk, while treating statistical significance as clinical significance.

A systematic review published 12 September 2026 searched three databases to 1 January 2025 and included 29 studies that directly compared machine learning CVD risk predictions with the Framingham Risk Score in healthy adults. Twenty-three studies reported improved predictive ability, often by adding sociodemographic predictors absent from FRS or costly diagnostics such as CT angiography.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
95

Updated Sep 14, 2026 · TRV-2026-1081

68Index score

Recomputed live from the record · Oct 12, 2026, 4:26 AM