Human Contractors Reviewing Content for Microsoft Copilot
According to a report from 404 Media, human contractors are reviewing prompts and images submitted to Microsoft's AI assistant, Copilot. These workers have encountered disturbing content such as explicit or potentially illegal material in their tasks. Internal messages reviewed by the outlet show that some contractors expressed concern over handling unsafe content without proper safeguards. One contractor described encountering a set of eight upskirt photos, while another mentioned seeing images related to an animal sacrifice. The report highlights growing concerns around AI moderation and ethical oversight at major tech firms like Microsoft. This issue raises questions about the safety protocols in place for human reviewers working with generative AI systems and could impact public perception of Copilot's reliability and Microsoft’s commitment to responsible AI practices.
Tickers: $MSFT