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

477
HealthStableModerate evidence · 1 source

A CNN with contrastive learning and mixed-attention fusion of multimodal MRI and demographic data achieved test set ACC 0.855 and AUC 0.919 for preoperative identification of pineal region germinoma, outperforming single-modality models.

Researchers developed a multimodal MRI-based deep learning model to non-invasively identify germinomas among pineal region tumors before surgery. Using 114 pathologically confirmed cases divided into training and test sets, a CNN with contrastive learning and a mixed-attention fusion of MRI sequences and demographic data was evaluated against single-modality models.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
89

Updated Aug 18, 2026 · TRV-2026-0825

67Index score
478
HealthStableModerate evidence · 1 source

Automated fundus-image analysis for diabetic retinopathy enables earlier screening and classification that can prevent vision loss in diabetic patients.

On 2026-08-17, a review in Graefe's Archive summarized automated diabetic retinopathy detection and classification using fundus images, surveying ML, DL and hybrid methods, their datasets, pre-processing and metrics, and noting newer ensemble, transformer and attention approaches.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
89

Updated Aug 18, 2026 · TRV-2026-0824

67Index score
479
HealthStableModerate evidence · 1 source

Psychiatric nursing internship students who used the ChatGPT-based virtual patient with a structured prompting framework showed higher MSE competency after the intervention.

Researchers tested an AI-powered virtual patient application built on ChatGPT to teach Mental Status Examination skills to 27 psychiatric nursing internship students. Using a sequential explanatory mixed-methods design, they measured competency before and after the structured prompting intervention and then interviewed 18 students about their experience.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
89

Updated Aug 18, 2026 · TRV-2026-0820

67Index score
480
ScienceStableModerate evidence · 1 source

A hybrid quantum-classical model combining variational quantum circuits with ESM2 achieved statistically significant improvements over classical baselines for multi-class IDR binding partner prediction.

Researchers developed a hybrid quantum-classical machine learning approach that pairs variational quantum circuits with the ESM2 protein language model in a prototypical network to predict binding partners of Intrinsically Disordered Regions across multiple classes including proteins, nucleic acids, lipids, and metal ions.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
89

Updated Aug 18, 2026 · TRV-2026-0819

67Index score

AI problems · 772

477
EducationStableModerate evidence · 1 source

Generative AI in education brings limitations, potential disruptions, ethical consequences, and risks of misuse if adopted hastily without efficacy and ethics review.

On September 2, 2024, a peer-reviewed commentary in Behaviour & Information Technology brought together nine experts to assess generative AI and large language models in education. The authors noted capabilities to enhance learning design, regulation of learning, and automated content, feedback, and assessment, while also flagging limitations, disruptions, ethical consequences, and misuses.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0358

67Index score
478
HealthStableModerate evidence · 1 source

Human evaluation practices for LLMs in healthcare show gaps in reliability, generalizability, and applicability, undermining assurance of safety and effectiveness.

Published September 28, 2024, this peer-reviewed review in npj Digital Medicine examined 142 studies of human evaluation methods for large language models in healthcare. The authors assessed how evaluations are designed and conducted, including dimensions evaluated, sample characteristics, evaluator recruitment, metrics, and analysis, and found systematic gaps in reliability, generalizability, and applicability.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0355

67Index score
479
PolicyStableModerate evidence · 1 source

When deployed in criminal justice, social welfare, fraud detection and public health, standard ML models can produce unreliable and harmful predictions due to misalignment with public-sector realities.

This peer-reviewed paper from October 2024 examines why AI-driven decision-making systems becoming instrumental in the public sector often fail to translate predictive accuracy into better decisions, analyzing five challenges including distribution shifts, label bias, past decisions shaping data, competing objectives, and human-in-the-loop effects.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0354

67Index score
480
SportsStableModerate evidence · 1 source

AI coaching systems risk privacy violations that expose sensitive athlete data, biased training algorithms that distort competitive fairness, and unclear responsibility for failures.

Published March 30, 2026, this peer-reviewed examination describes AI coaches that create customized, data-driven training programs to optimize athletic performance, while warning that privacy breaches, biased algorithms, and unclear accountability threaten personal rights and fairness in competition.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
84

Updated Jul 20, 2026 · TRV-2026-0351

67Index score

Recomputed live from the record · Oct 13, 2026, 12:25 PM