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

325
HealthStableModerate evidence · 1 source

Adolescents with higher rumination and depression showed greater authentic self-disclosure to generative AI, indicating GenAI is being used as an outlet for personal disclosure.

A peer-reviewed study published September 23, 2026 surveyed 685 Chinese adolescents averaging 13.17 years old to examine how parental attachment anxiety relates to authentic self-disclosure to generative AI. It found father and mother attachment anxiety were positively associated with rumination, which was linked to depression, and both rumination and depression were positively associated with disclosure to GenAI through sequential mediation.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
97

Updated Sep 27, 2026 · TRV-2026-1199

68Index score
326
EducationStableModerate evidence · 1 source

Instructors using generative AI to draft fillable study guides for introductory biology can reduce drafting time and get help organizing content and creating alternative question formats.

Published September 23, 2026 in Journal of Microbiology & Biology Education, this Tips and Tools article outlines how an instructor can use generative AI to draft fillable study guides for introductory biology. The author provides lecture content, assignments, and example worksheets as context for an initial prompt, then iteratively refines the AI draft for accuracy and alignment.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
97

Updated Sep 27, 2026 · TRV-2026-1198

68Index score
327
ScienceStableModerate evidence · 1 source

By October 2023 testing, six generative chatbots including GPT-4 produced ideas judged equal in quality to human ideas, indicating they can function as assistants in creative work.

In a study published October 9, 2023, researchers tested the view that AI cannot be creative by having humans and six generative chatbots generate ideas, then having humans and a trained AI evaluate them. The evaluation found no qualitative difference between the two sources on average.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
97

Updated Sep 26, 2026 · TRV-2026-1192

68Index score
328
LifestyleStableModerate evidence · 1 source

In contrast, the combination of Whale Optimization Algorithm (WOA) and PLS achieved the highest validation estimation accuracy (R 2 = 0.7452).

Real-time and accurate monitoring of heavy metal concentrations in rice is essential for ensuring food safety and supporting the safe utilization of contaminated agricultural land. Hyperspectral technology offers advantages such as rapid, nondestructive, cost-effective, and environmentally friendly monitoring.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
97

Updated Sep 26, 2026 · TRV-2026-1188

68Index score

AI problems · 771

325
HealthStableModerate evidence · 1 source

Most AI models for predicting laparoscopic cholecystectomy difficulty carry high risk of bias, rarely undergo external validation, and have significant methodological flaws limiting clinical translation.

By August 13, 2026, a systematic review and meta-analysis of 18 studies found AI models predicted laparoscopic cholecystectomy difficulty with pooled AUCs of 0.848 in training and 0.818 in validation, with ensemble models reaching 0.889 and 0.861. The review searched four databases to March 2, 2026 and used PROBAST and GRADE to assess bias and certainty.

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

Updated Aug 16, 2026 · TRV-2026-0787

67Index score
326
HealthStableModerate evidence · 1 source

Labor-intensive segmentation and factors like comminution, bilateral injury, loss of anatomic references, and metal artifacts still restrict efficiency and reproducibility of digital reconstruction workflows.

This narrative review from August 2026 summarizes AI applications across occlusion-oriented digital reconstruction of maxillofacial fractures, where treatment must address stable occlusion, mandibular movement, temporomandibular joint position, facial contour, and fixation as interdependent targets. It evaluates tasks from CT/CBCT screening and segmentation to model repair, shape completion, planning assistance, and postoperative deviation analysis.

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

Updated Aug 16, 2026 · TRV-2026-0781

67Index score
327
HealthStableModerate evidence · 1 source

AI facial image scoring, editing and curation systems converge on a narrow westernized phenotype and are linked to appearance dissatisfaction, perception drift, and Snapchat and Zoom dysmorphia presentations in plastic surgery patients.

This educational review from August 2026 examines how AI systems that score, edit, generate and curate facial images encode explicit quantitative definitions of attractiveness and deliver them through prediction algorithms, AR filters, generative imagery and surgical simulators. It finds independent models converge on a narrow, frequently westernized phenotype and summarizes evidence linking such imagery to appearance dissatisfaction and perception drift.

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

Updated Aug 16, 2026 · TRV-2026-0780

67Index score
328
HealthStableModerate evidence · 1 source

Because the dataset was split at the image level, correlated frames from the same examination may inflate performance, so results cannot be interpreted as patient-level generalization and require grouped reanalysis and external validation before clinical use.

A comparative study merged SEE-AI and Kvasir-Capsule into a 21-class capsule endoscopy image dataset and fine-tuned a Vision Transformer, DenseNet121, and ResNet50. On an independent test set of 8,696 frames, the transformer achieved 92.2% accuracy and 0.99 AUC, substantially higher than the two CNN baselines under the reported experimental conditions.

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

Updated Aug 16, 2026 · TRV-2026-0779

67Index score

Recomputed live from the record · Oct 12, 2026, 5:58 PM