GoKawiilOpenAI contractors fired for using AI to help train ChatGPT
404 Media reports that contractors hired to review and critique real ChatGPT conversations to improve OpenAI's models have been terminated for secretly using AI tools to generate their feedback instead of doing it themselves. Multiple contractors told the outlet this is a widespread and frequently punished practice among the thousands of workers doing this human-review work.
The human review process exists precisely to prevent AI models from degrading through 'model collapse,' where systems trained on AI-generated content get progressively worse. If the humans meant to safeguard quality are quietly outsourcing that judgment back to AI, it undermines the entire premise of human-in-the-loop training and raises questions about the actual quality of feedback shaping widely used chatbots.
- OpenAI contractors reviewing real ChatGPT conversations have been fired for using AI to generate their critiques.
- The practice appears common, with workers describing frequent firings across a workforce of thousands.
- The issue highlights risks of 'model collapse' when AI-generated content contaminates the human feedback meant to improve AI systems.
Source: news.ycombinator.com — About The Author, 2026-09-22
Published there as: “People Training OpenAI's AI Fired for Using AI to Train the AI”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.