Ilyas Mutlu
Author at Caywork
In the rise of cheap writers and non-complainers (A.I.), humans started to create an infinite information machine for low-quality articles. The information or training data of A.I. is not infinite, but we treat it as if it is when we say things like "create me an article about 'Why A.I. is so good?'" It creates the article like a writer, but there is a good chance that when you give same input enough, you will see the pattern. And you are far from the only one giving the same input.
This results in A.I. content being the same when published by everyone. The problem is not just that everyone is publishing the same thing, but also that hallucinating A.I. is not uncommon. A.I. may make up facts or provide inaccurate information, posing a significant threat of junk A.I. ruining the internet.
In the end, A.I. is not an evil toy to play with; you can use it to output different views from yourself if you create it with the original idea, or you can fix spelling or simplify it. The tool is just a tool. What differs is how you use it.