OpenAI claimed it's "impossible" to build good AI models without using copyrighted data. An “ethically created” large language model and a giant AI dataset of public domain text suggest otherwise.
Read this story here.
Learn more about your ad choices. Visit megaphone.fm/adchoices
