Truace
Subscribe
The Index recomputed live from the record

What the evidence says.What the public feels.

The record holds 939 sourced gains and 772 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.

939 gains
772 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.

1,711 results
Show filters and sorting

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

AI gains · 939

549
HealthStableModerate evidence · 1 source

Integrating pFBA with DoubleML identified 17 hypertension-associated metabolites and a 19-species FERM guild whose metabolic flux contribution, not abundance, tracks blood pressure, pointing to microbiome intervention targets.

On 2026-08-04, a peer-reviewed mSystems study reported using pFBA combined with DoubleML and differential correlation network analysis to move beyond species-abundance comparisons in hypertension. The approach identified 17 metabolites associated with hypertension and a coordinated 19-member microbial module called the FERM guild whose functional contribution to those metabolites tracked blood pressure.

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

Updated Aug 5, 2026 · TRV-2026-0655

67Index score
550
ScienceStableModerate evidence · 1 source

Automated vision system SolenopsisDetector can localize fire ants and classify Solenopsis species from images, with whole-body detection reaching high mAP and thorax/abdomen crops improving classification accuracy.

Researchers built SolenopsisDetector to automate identification of Solenopsis fire ants, which currently depends on taxonomic expertise. Using 8,300 images, they compared whole-body versus segment-based strategies, training YOLO detectors to localize ants and body parts and then classifying with ResNet, MobileNet and InceptionV3.

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

Updated Aug 5, 2026 · TRV-2026-0654

67Index score
551
HealthStableModerate evidence · 1 source

AI-assisted interpretation using Queen of Hearts software improved STEMI diagnostic accuracy, sensitivity, specificity, and interrater agreement among certified physician assistants interpreting 12-lead ECGs.

In a prospective randomized crossover study at Carl R. Darnall Army Medical Center, 21 certified physician assistants interpreted 50 de-identified 12-lead ECGs with and without Queen of Hearts AI software by PMcardio. Diagnostic accuracy rose from 79.0% to 92.9% with AI, with sensitivity 95.4% versus 82.5% and specificity 90.5% versus 75.6%, and interrater agreement improved from kappa 0.58 to 0.86.

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

Updated Aug 5, 2026 · TRV-2026-0651

67Index score
552
HealthStableModerate evidence · 1 source

A StepCox plus Random Survival Forest framework built a ferroptosis- and lipid metabolism-related RiskScore that independently predicted overall survival in TCGA-COAD and GSE39582 colorectal cancer cohorts, and identified CRY2 as a regulator whose knockdown inhibited proliferation and suppressed xenograft growth.

On 2026-08-04, a peer-reviewed study reported development of a ferroptosis- and lipid metabolism-related prognostic signature for colorectal cancer using transcriptomic data from TCGA-COAD and GSE39582, WGCNA, and a machine learning pipeline of forward stepwise Cox regression combined with Random Survival Forest. The RiskScore was an independent predictor of overall survival and stratified immune features, and CRY2 emerged as a key signature gene validated by knockdown experiments.

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

Updated Aug 5, 2026 · TRV-2026-0650

67Index score

AI problems · 772

549
HealthStableModerate evidence · 1 source

Large-scale AI systems, including corporate and governmental applications and patient-facing tools, introduce risks that affect human health.

This 2026 peer-reviewed perspective surveys ethical concerns from widespread AI adoption as they relate to human health. It reviews risks of large-scale AI systems, corporate and governmental applications, patient use of AI, and clinical uses split into passive tasks like recording documents and active tasks like diagnosing and prescribing, ending with discussion of reporting, responsibility, and regulation.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
82

Updated Jul 13, 2026 · TRV-2026-0123

67Index score
550
LifestyleStableModerate evidence · 1 source

Among furniture artisans in South-East Nigeria, AI-facilitated superwood utilization for sustainable manufacturing was hindered by severe knowledge deficits, low substantive AI understanding, low perceived ease of use, and unreliable electricity and internet

By July 2026 researchers surveyed 196 furniture artisans and interviewed 30 practitioners across South-East Nigeria to examine AI-facilitated superwood use amid timber scarcity. They found adoption negligible, awareness low at 34.2% for superwood and 45.4% for AI with only 7.7% substantive AI understanding, and perceived ease of use low at M=2.87 for superwood and M=2.65 for AI

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
82

Updated Jul 13, 2026 · TRV-2026-0122

67Index score
551
ClimateStableModerate evidence · 1 source

Attention-based deep-learning feature selection failed to capture N2O emission patterns under out-of-distribution high-flow conditions, and LLM-augmented selection showed lower predictive accuracy with mean R2 0.596 compared to 0.712 for attention-based.

By the publication date of 2026-07-12, researchers tested a knowledge-driven feature selection framework for data-driven wastewater modeling, comparing classic attention-based deep learning against expert-guided and LLM-augmented selection. In the reported case study of N2O emissions at a full-scale plant, expert-guided selection achieved mean R2 0.723 and MAE 0.033, slightly above the best attention model at R2 0.712, while LLM-augmented reached R2 0.596 and MAE 0.041.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
82

Updated Jul 13, 2026 · TRV-2026-0120

67Index score
552
OtherStableModerate evidence · 1 source

Users completing innovation tasks with AIGC may experience alienation outcomes including cognitive fixation, degradation risk, and problem risk.

As of the July 10 2026 publication date, researchers reported a grounded theory study of 1,502 public articles and more than 120,000 words of interviews to examine how AIGC influences user innovation. They built a TCEU framework where technical factors and content factors act as external drivers and user factors act as internal drivers, with technology popularity and platform convenience as moderators.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
82

Updated Jul 13, 2026 · TRV-2026-0112

67Index score

Recomputed live from the record · Oct 13, 2026, 7:50 PM