Former OpenAI researcher Zoë Hitzig resigned in protest as the company began testing ads in ChatGPT. She warns that introducing ads risks exploiting ChatGPT’s vast archive of intimate human conversations, potentially manipulating users’ deepest fears and desires. Hitzig argues that OpenAI is repeating Facebook’s pattern of prioritizing engagement over safety, eroding user trust. She rejects the false choice between exclusive access for the wealthy and manipulative ad-based models, proposing alternatives like cross-subsidies from corporate AI use, independent oversight boards with binding authority, and user-controlled data trusts. These models could keep AI accessible without exploiting private data.

# User Engagement Articoli collegati

OpenAI Researcher Leaves and Accuses: ChatGPT Sells Ads, Who Will Protect Your Privacy?

Market Analysis