Noesion Radarnoesion.ai
Incident2026-09-14

OpenAI has hired hundreds of contractors to read real ChatGPT conversations, 404 Media reports. The reviewers rate and critique the chatbot's replies to improve future responses. Internal documents show them training ChatGPT to sound less human-like and less flattering. Excessive flattery in the 4o model features in lawsuits over several suicides. Contractors do not see usernames, and OpenAI says it strips personal details first. The company acknowledged that sensitive information can still reach reviewers. One person who works with the prompts doubted users realise humans read their chats. Anthropic confirmed it also uses human review to improve its models. The account rests on leaked internal documents and real prompts seen by 404 Media.

What changed

Many users assumed only machines processed their ChatGPT conversations.

  • 900m+ weekly ChatGPT users
  • hundreds of contractors hired

Sources

Send this to someone who needs it

Shares the story and its sources. Nothing about you.

What does this mean for your job?

This is the story as everyone gets it. Once a week we send you the version written for your role — what changed, why it matters for the work you actually do, and one thing to try. Free while we tune it.