# Complaint rates rose 30% for same models after controlling for task mix

> Original title: Part of the rise came from what users asked for. In June and July, more requests were for full apps, which draw more complaints than sing...

- Source: Design Arena
- Published: 2026-10-30T16:36:00.000Z
- Bad Influence: https://badinfluence.news/items/mwle0djulhoab6sffx0dd1rqj
- Original: https://x.com/DesignArena/status/2116207501997916536

## Summary

Even after holding task type, language, chat position, user tenure and request length fixed, the same models drew about 30% more complaints, rising from 4.2% to 5.5% of replies. The analysis regressed complaint rates on indicator variables for each model and each control factor, and the 30% figure is the rise left after those effects are removed. Part of the overall increase came from users asking for full apps more often in June and July, which draw more complaints than single web pages.

## Post

Part of the rise came from what users asked for. In June and July, more requests were for full apps, which draw more complaints than single web pages.

But with the task, language, point in the chat, how long each user had been using the chat, and request length all held fixed, the same models still drew about 30% more complaints, from 4.2% of replies to 5.5%.

To measure this, we regressed against an indicator variable for each model and for each category of those factors, which estimates how much each one raises or lowers the complaint rate. The 30% is the rise left after taking those effects out.

[Video](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Famplify_video_thumb%2F2116021243787051008%2Fimg%2F2o__WkI43plD3Jjs.jpg&mode=thumb&exp=1793577600&sig=0c425640aedb5596)