In early prototype testing, AI-generated visuals were the main barrier to real-world readiness, rated lower than text and narration¶
Claim · All claims
Evidence · 1 study · designr3·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Phase 2 practitioners rated the quality and consistency of AI-generated visuals lower than text and narration, citing wait times, generation failures, and cross-scene inconsistencies. → Buyankhishig Enkhjargal 2026
Evidence¶
Buyankhishig Enkhjargal 2026¶
Buyankhishig Enkhjargal, Himanshi Lalwani, Hanan Salam. (2026). AI-Assisted Social Story Intervention for Special Education: The Design of AdaptED Stories. arXiv preprint. https://arxiv.org/abs/2609.24245 doi:10.48550/arxiv.2609.24245
q2 · i? · design · r3
Phase 2 prototype evaluation with three practitioners using SUS and 5-point Likert items; visual quality received a mean of 3.3 in Figure 11 while the average SUS was 76.6. Qualitative complaints named protagonist appearance changes and artistic style variation.
"Compared with text and narration, practitioners gave lower ratings to the quality and consistency of AI-generated visuals and described long wait times, occasional generation failures, and inconsistencies across scenes, such as changes in the protagonist's appearance and variations in artistic style."
Discussion¶
Related Claims¶
- Special-education practitioners rated AdaptED Stories as strongly usable, with a mean SUS score of 86.8 corresponding to an A grade — related
- Practitioners perceived the profile-driven workflow as reducing Social Story preparation burden — related
- Students identified weaknesses in the AI feedback model: lack of real-time feedback, software/economic access barriers, and occasional incorrect or inconsistent outputs — related
- Reviewer free-text comments identified synthetic voice, slide–narration reveal timing, and slide layout as the main remaining quality issues — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
ai-visual-generation-main-barrier-prototype |
title |
In early prototype testing, AI-generated visuals were the main barrier to real-world readiness, rated lower than text and narration |
description |
In early prototype testing, AI-generated visuals were the main barrier to real-world readiness, rated lower than text and narration |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — buyankhishig-enkhjargal-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.