Updated

Which AI model writes the most accessibility and UX issues?

As of Sep 27, 2026, Claude Sonnet 5 had the highest rate of accessibility and UX issues, 54% above the all-model rate (1.54), statistically tied with 4 other models. GPT-5.5 had the lowest, 51% below average (0.49), statistically tied with Claude Fable 5 and GPT-5.6 Sol.

Models ranked by relative rate of accessibility and UX issues, fewest first. Lower values are better.
RankModelRate
1= (statistically tied)GPT-5.5
0.49
2= (statistically tied)Claude Fable 5
0.67
3= (statistically tied)GPT-5.6 Sol
0.70
4= (statistically tied)Claude Opus 4.8
1.03
5= (statistically tied)GPT-6 Astra
1.08
6= (statistically tied)Claude Fable 5.1
1.21
7= (statistically tied)Claude Opus 5
1.40
8= (statistically tied)Claude Sonnet 5
1.54

1.00 = all-model rate · lower is better · = statistically tied