Yeah... I mean the pre-trained text encoder Firefly uses is filled with copyrighted, unlicensed-for-express-purpose training data.
Firefly is successful in the same sense that DocuSign is, in that it is selling a holistic social experience, but I think maybe don't opine on "legal provenance on their training data" until you have seen the whole pipeline with your own eyes. Sophisticated people sort of know Adobe's claims are bullshit, but what exactly do you expect the community to do, speculate on the exactly zero evidence Adobe has shown of how any of their stuff works?
Anyway, I am pretty sure Adobe is delighted they are not paying $20b for something that is worth way, way less. Like maybe $500m at most.
Meanwhile the people using Figma at many companies are getting laid off. Etsy, Bytedance, Unity Spotify, Salesforce all made massive UX designer cuts.
The real question is, is the thing people are using Figma for even worth $20b? No, no way. Figma users work in the Making Bugs department: they make new buggy things nobody asked for, that aren't lists or spreadsheets but should just be lists and spreadsheets, which makes everything worse. There is nowadays positive ROI to doing less Figmaing. In my opinion there has always been more ROI to doing less Figmaing, to straight up not having those people around and not gathering so many opinions on designing lists from so many stakeholders. That holistic experience is expensive in many ways, and while again you can be successful delivering that, it doesn't mean it makes sense.
Just look at the Spotify app. It's a hot abject mess of absolute garbage UI. They have been diehard Figma users for years. They are the prime example of Figmafication ruining something extremely simple. It's fucking lists! Lists are not worth $20b.
