... plumbing, circuits, etc and complete stress testing. Comment from Nikita Lokhmachev(@nikita_lokhmachev): the sandboxing framing here is interesting. apple's concern isn't really about vibe coding per se, it's about runtime code execution escaping the review perimeter. the "self-contained app" rule exists because the trust model assumes you can audit what ships. once an app can generate and run arbitrary logic at runtime, that audit becomes meaningless. the app you reviewed isn't the app running ... ... different from installing a private APK instead of downloading from an app store, where users expect vetted apps. Comment from Rohan Chaubey(@rohanrecommends): @emmanuel_afolabi  I like how you framed it as speed of creation vs

... agents that understand their own confidence level and ask for approval only when the stakes are high. We are not there yet, but the trajectory is encouraging. Comment from Henrik Pedersen(@henrikpedersen): @busmark_w_nika  Exactly passive vs active is a cleaner way to put it. The moment AI acts as you rather than for you, the trust equation changes completely. Comment from Ben Sabic(@bensabic): Like with most people, I'd trust AI for most things, except ... ... things you listed (e.g. personal finances). At least not without approval processes built in (e.g. get a push notification to approve paying a bill). Comment from Henrik Pedersen(@henrikpedersen): The distinction I keep coming back to is read vs