Is $1B enough to license or create all the training data said models and ais will need? Data sets aren't quite as easy to scrape and copyright infringe on as t

Is $1B enough to license or create all the training data said models and ais will need? Data sets aren't quite as easy to scrape and copyright infringe on as they were before chatGPT