cross-posted from: https://piefed.social/c/technology@piefed.social/p/2533091/proton-to-re-interview-staff-about-ai-to-see-who-gets-to-keep-their-job
cross-posted from: https://piefed.social/c/technology@piefed.social/p/2533091/proton-to-re-interview-staff-about-ai-to-see-who-gets-to-keep-their-job
For me personally, I have not used any AI product or service and trying my utmost to avoid any product or service where ‘AI’ scans my data. I use donotspy on my windows to stop all the shit that now comes with Microsoft as unfortunately my work requires a windows platform, Firefox because I can disable all that AI shit and have my extensions up and running.
It is nothing more than an advanced version of ‘if… then… else’ except it gets it wrong 48% of the time.
Today I had to find a new email client because EMclient also decided to add slop to their app.
As much as I appreciate the warning about uploading any file to any AI client, never in my fucking life.
Yeah the problem is any AI that is built ethically and respects privacy even 48% right is being kind. While the unethical ones are starting to push 90%+ accuracy in god damn well near every metric the normal person cares about.
But they abuse the Internet and destroy everything to get there.
It’s fucked
Smart. Good on you for being mindful and taking appropriate precautions.
I have a Windows/Linux dual-boot and keep anything requiring privacy on the Linux side. For Windows, I use O&O ShutUp, which is an application that allows a user to easily shut off all of Windows’ intrusive settings and delete Copilot.
As for LLM’s, they’re my last resort for anything, and I only really use them if I can’t find the information I need on Mojeek or Marginalia-Search, not just for privacy’s sake but because I have an ideological problem with their vast energy use and the economic and environmental impact those applications have.
But none of it’s foolproof, so I always just keep in the back of my mind before doing or saying anything online: Would I want this to go public?