In the eyes of big AI companies such as OpenAI, the troves of data on the internet are highly valuable. They scrape photos, videos, books, blog posts, albums, painting, photographs and much more to train their products such as ChatGPT – usually without any compensation to or consent from the creators.
In fact, OpenAI and Google are arguing that a part of American copyright law, known as the “fair use doctrine”, legitimises this data theft. Ironically, OpenAI has also accused other AI giants of data scraping “its” intellectual property.
