Lawyers are funny. Saying ChatGPT does not plagiarize “Willy nilly” is a cute way of saying “but it *does* plagiarize.”
I could open MS Word infinity times, and it will never accidentally open NYT articles.
I could open Google Docs infinity times, and it will accidentally plagiarize NYT.
I could open MS Word infinity times, and it will never accidentally open NYT articles.
I could open Google Docs infinity times, and it will accidentally plagiarize NYT.
Reposted from
Kate Knibbs
OpenAI lawyer claiming he thinks NYT had to try thousands or tens of thousands of times to prompt ChatGPT to regurgitate its articles... says it won't otherwise do it "willy-nilly"
Comments
If we give them grace and say every 100,000 words, it plagiarizes, I should be able to find 1,120 instances of plagiarism since 2011.
I have 0 instances to report.
“I don’t rob houses Willy nilly. There’s 1 million houses in this city. I only robbed one, and it took effort.”
“I don’t murder people Willy nilly. There’s 2 million people living here. I only killed one, and I had to try.”
Do these people even have the common sense to know that they are complete dumb and not "geniuses and that others don't understand AI"?