3 ms·

> ur chatGPT phone would be used as your 2nd brain / your AI friend. That's what we need, a brain/"friend" that collects and analyzes every single thing we th

> ur chatGPT phone would be used as your 2nd brain / your AI friend. That's what we need, a brain/"friend" that collects and analyzes every single thing we think/ask about to use against us later and/or sell to third parties. No thank you! The only AI "friend" you'll ever have will be run locally entirely on your own hardware and will never send/share/leak any part of your interactions with anyone else for any reason.

pessimizer 2y ago
Any AI personal assistant that is working for me should do everything short of illegality to protect any information about me, or what I do, that I haven't explicitly told it I want to be shared, and with whom. You might say that's crazy, but that's just Overton. It's the expectation we have of librarians (and that they have of themselves.) It took PATRIOT to erode that norm (and many others.) Right now, that at the least means running it locally, although that's not going to be sufficient.
- rchaud 2y ago
  So, what, you're going to get the AI to sign an NDA? They don't work for you, they work for the vendor. You can't hold an AI accountable, so you shouldn't trust it with your secrets.
paul7986 2y ago
Maybe time will tell (im betting yes on AI powered phone as noted above and via chatGPT/Microsoft) and there's two camps currently .... like myself i sing the praises of Meta RayBan's smartglasses ... i can take pics/videos way easier then pulling out my phone. Yet if i pull out my phone one isnt going to know 100% if they are being video(ed) or photographed unless it's very obvious (they are following the subject with their phone held up). Meta's Smartglasses makes it less obvious and i guess those in the camp against meta's glasses would argue it erodes privacy. Yet legally in public spaces we have zero privacy! I guess it's a control thing.

3 ms·

> ur chatGPT phone would be used as your 2nd brain / your AI friend. That's what we need, a brain/"friend" that collects and analyzes every single thing we th

by autoexec 2y ago

pessimizer 2y ago
Any AI personal assistant that is working for me should do everything short of illegality to protect any information about me, or what I do, that I haven't explicitly told it I want to be shared, and with whom. You might say that's crazy, but that's just Overton. It's the expectation we have of librarians (and that they have of themselves.) It took PATRIOT to erode that norm (and many others.) Right now, that at the least means running it locally, although that's not going to be sufficient.
- rchaud 2y ago
  So, what, you're going to get the AI to sign an NDA? They don't work for you, they work for the vendor. You can't hold an AI accountable, so you shouldn't trust it with your secrets.
paul7986 2y ago
Maybe time will tell (im betting yes on AI powered phone as noted above and via chatGPT/Microsoft) and there's two camps currently .... like myself i sing the praises of Meta RayBan's smartglasses ... i can take pics/videos way easier then pulling out my phone. Yet if i pull out my phone one isnt going to know 100% if they are being video(ed) or photographed unless it's very obvious (they are following the subject with their phone held up). Meta's Smartglasses makes it less obvious and i guess those in the camp against meta's glasses would argue it erodes privacy. Yet legally in public spaces we have zero privacy! I guess it's a control thing.