Adobe FREAK OUTS! Addresses TOS Backlash Dumpster Fire!
(www.youtube.com)
You're viewing a single comment thread. View all comments, or full comment thread.
Comments (22)
sorted by:
I'm still confused on the reasoning for Adobe to monitor customer content. If they need to monitor content uploaded to their server or generated using their generative AI feature, then I could give them the benefit of the doubt for that. But if I'm understanding the terms of service correctly, then they're asking for access to all content created with or imported into their software. Aren't their tools supposed to work without an internet connection?
"We may access, view, or listen to your Content"
'"Content" means any text, information'...'images, that you upload, import into, embed for use by, or create using the Services and Software'
Oh really? I thought people were still cracking it.
I think that's the version I have, that used to work, but it didn't work when I went back to it a while ago.
I'll be as clear as possible.
They need image data to feed new, enormous training sets for neural networks. they need as much data as possible and they need it right now. there is a data gold rush that's been going down for about 10 years now (2015, major breakthrus in NNets), people realize this now, and so they are walling off data sources they control and/or charging money.
Adobe is integrating NN tools into their products like context fill and such. these tools require MASSIVE amount of data for their training sets.
I did think it was interesting that Adobe clarified that they don't train the Firefly AI model on customer content. My first thought when reading that was, "so you're training other AI models?"