Oracle's Larry Ellison argues training on public data commoditizes AI, leaving proprietary datasets as the only moat

i warned these guys of exactly this problem - no moat because everyone is training on same data - two years ago. and warned them that AI would become a commodity. now they think they have made some big discovery. 🤣 the inability of these guys to listen to anyone outside of Silicon Valley is gonna cost them a truly enormous amount of money. (and no proprietary data is not going to make that much of a difference, except in some narrow use cases)