OpenAI Hires Contractors to Read Chats for Model Development
The company aims to improve artificial intelligence models and enhance responses by reviewing ChatGPT interactions with user consent.
OpenAI has started hiring hundreds of contractors to read chats with user consent in order to improve artificial intelligence models and make ChatGPT responses better.
New Teams Being Formed to Read Messages
According to internal correspondence and emerging information, OpenAI is hiring hundreds of contractors to review ChatGPT messages in cases where users grant permission.
Model Development and Response Quality Goal
The messages reviewed by these personnel cover conversations between the user and the chatbot, with the goal of improving the responses that ChatGPT provides to users.
Privacy Risks and Sensitive Information
Because millions of people use ChatGPT like a therapist or friend and share private information, there is a risk that identities could be exposed from conversation contents even if names are masked.
Permission Settings and Consent Mechanism
When users enable the 'Improve the model for everyone' setting on the ChatGPT platform, they agree to have their conversations read by teams hired by OpenAI.