Truace
Subscribe
The Index recomputed live from the record

What the evidence says.What the public feels.

The record holds 935 sourced gains and 771 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.

935 gains
771 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.

1,706 results
Show filters and sorting

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

AI gains · 935

373
LifestyleStableModerate evidence · 1 source

In 230 SMEs, green innovation was positively related to sustainable performance, mediated by green knowledge sharing and green dynamic capabilities, with AI significantly moderating those links.

Researchers examined how green innovation relates to sustainable performance in small- and medium-sized enterprises, testing green knowledge sharing and green dynamic capabilities as mediators and artificial intelligence as a moderator using two-phase data from 230 SMEs.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
94

Updated Sep 9, 2026 · TRV-2026-1037

68Index score
374
EducationStableModerate evidence · 1 source

In medical education, AI-enabled tools can improve training efficiency by providing faster workflows and richer learning resources that strengthen training.

A 2026 peer-reviewed paper analyzes artificial intelligence in medical education, noting that AI is rapidly changing training by offering faster workflows and richer learning resources, while also quietly reshaping how future clinicians think and act.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
94

Updated Sep 9, 2026 · TRV-2026-1036

68Index score
375
HealthStableModerate evidence · 1 source

Protocol proposes AI-based DyVe-X software to continuously quantify exertional dyspnoea against work rate and ventilation and to identify excessive and constrained breathing patterns during incremental CPET, with anticipated superior performance over peak breathing reserve criterion.

Published 2026-09-07 as a peer-reviewed protocol, the study outlines validation of Dynamic Assessment of Dyspnoea and Ventilation on Exercise (DyVe-X), an AI-based software that continuously assesses dyspnoea intensity and mechanical-ventilatory reserve depletion during incremental cardiopulmonary exercise testing in 1161 tobacco-exposed subjects.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
94

Updated Sep 9, 2026 · TRV-2026-1033

68Index score
376
HealthStableModerate evidence · 1 source

Data augmentation and feature selection techniques may improve robustness, predictive performance, and interpretability of machine learning models for autism spectrum disorder prediction and help address dataset scarcity.

This peer-reviewed review examined 26 studies from 2021 to 2024 on machine learning and deep learning for autism spectrum disorder prediction, focusing on data augmentation and feature selection methods. It categorized augmentation into conventional transformations and GAN-based synthetic generation, and feature selection into filter, wrapper, and embedded approaches, evaluating study quality with the Prediction Model Risk of Bias Assessment Tool.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
94

Updated Sep 9, 2026 · TRV-2026-1032

68Index score

AI problems · 771

373
PolicyStableModerate evidence · 1 source

Human Digital Remains created by AI from personal and biometric data face existing legal and ethical gaps because neither GDPR nor the AI Act currently extends rights to the deceased.

The paper examines how AI and extended reality enable creation of avatars and human digital twins from personal and biometric data that persist after death as Human Digital Remains. Using cross-disciplinary analysis and doctrinal review, it finds that current EU instruments do not extend protections to the deceased and identifies urgent legal and ethical gaps.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
86

Updated Aug 1, 2026 · TRV-2026-0620

67Index score
374
HealthStableModerate evidence · 1 source

Inefficient low-power field target detection, reflected in longer fixation duration on the LPF main object, predicts lower diagnostic accuracy, while traditional years of professional experience fails to predict accuracy in digital cytology.

Researchers eye-tracked 100 cytotechnologists diagnosing 30 digital cytology images and then tracked 28 students before and after a 3-month training program. They found years of experience did not predict accuracy, while shorter fixation on the low-power field main object did, and students markedly improved time to first target fixation and reduced background attention after training.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
86

Updated Aug 1, 2026 · TRV-2026-0619

67Index score
375
HealthStableModerate evidence · 1 source

There are few controlled clinical trials that directly compare AI tools to traditional medical literature and clinical experience for hypertension on key endpoints of real clinical value to prove superiority.

By August 2026, peer-reviewed discussion in Giornale Italiano di Cardiologia described AI as entering hypertension care, able to give simple and well-documented answers to management questions for practicing physicians, while also being explored in research to identify secondary hypertension and predict future hypertension and complications such as heart failure.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
86

Updated Aug 1, 2026 · TRV-2026-0617

67Index score
376
HealthStableModerate evidence · 1 source

Responses showed comparatively lower inclusivity, high reading complexity, and lacked nuance for complex or individualized clinical scenarios.

In this cross-sectional study published August 1 2026, researchers asked GPT-4 via ChatGPT to answer 20 common psychosis psychoeducation questions sourced from a first-episode psychosis programme, then had two psychosis experts independently rate the answers on accuracy, clarity, inclusivity, completeness, clinical utility and overall quality.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
86

Updated Aug 1, 2026 · TRV-2026-0613

67Index score

Recomputed live from the record · Oct 13, 2026, 12:54 AM