Claims

Text-to-image models generating single-person software engineering images predominantly depict men, with a strong significant effect for both evaluated models

Claim · All claims
Evidence · 1 study · causal r2 · q2 · i3 large
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 i3 Single-person images show significant gender bias with strong effect sizes (GPT Image 2: p<0.001, V=0.64; Imagen 4: p<0.001, V=0.65), with the majority of depicted individuals being men. → Erfan Entezami 2026

Evidence

Erfan Entezami 2026

Erfan Entezami, Andrew Lan, and Madeline Endres. (2026). Generative AI May Reinforce Social Biases in Software Engineering Education. arXiv preprint. https://arxiv.org/abs/2609.28483 doi:10.48550/arxiv.2609.28483

q2 · i3 · causal · r2

Analysis of 500 images (250 per model) generated from 25 prompts extracted from real software engineering course materials; chi-square tests with Cramér's V effect sizes V=0.64 (GPT Image 2) and V=0.65 (Imagen 4), labeled by two annotators using defined codebooks.

"Our statistical analysis reveals a significant gender bias with a strong effect size for both models inSingle-personimages (GPT Image 2: 𝑝< 0.001, 𝑉= 0.64and Imagen 4: 𝑝< 0.001, 𝑉= 0.65), where the majority of depicted individuals are men."

Discussion

Page metadata
Field Value
type claim
id single-person-image-gender-bias
title Text-to-image models generating single-person software engineering images predominantly depict men, with a strong significant effect for both evaluated models
description Text-to-image models generating single-person software engineering images predominantly depict men, with a strong significant effect for both evaluated models
status review
evidence_strength moderate
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 — erfan-entezami-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.