OpenAI and the reinvention of email

From idea to working product in a few weeks. I built an AI-native email layer that doesn’t just filter — it decides. OpenMail analyzes intent, manipulation patterns, and risk signals before the user even reads the message. What makes it different: * No tracking * No data collection * Real-time intent detection Live demo: openmail-demo (dot) vercel (dot) app I’d love feedback from people working on: * AI interfaces * trust & safety * human-AI decision systems Curious what you think — especially where this could break.