Art of Data Engineering reposted this
Maybe I'll regret writing this post one day, and I know this will divide opinion, but when it comes to AI my overarching feeling is that... Big to Small (i.e. condensing large amounts of data into a set of predictions, or using GenAI for summarisations/image analysis etc) = Overall positive Small to Big (i.e. transforming a small prompt into an essay, music track, video etc) = Overall negative That's not to say that there aren't good applications in the latter category, and of course there are edge cases in each, but in very general terms this feeling is becoming increasingly hard to shake.