At risk of being too pedantic even for HN, I have to say that a news article written by GPT-3 isn't "Fake News". Maybe you could called it "artificially authore

At risk of being too pedantic even for HN, I have to say that a news article written by GPT-3 isn't "Fake News". Maybe you could called it "artificially authored news" or something, but nothing about synthesizing and regurgitating words is inherently fake. "Fake News" is a loaded term that fact-checkers tend to avoid for its ambiguity. Generally its usage refers to disinformation, which is the use of media to intentionally deceive the reader for political or social motivations. Its terrifying to imagine artificially authored disinformation, but from my sparse understanding of GPT-3, it wouldn't be the right tool for crafting novel disinformation without a lot of inputs for its user. Disinformation is dangerous when represented as truth by platforms with credibility, and no content creation tool can garner and wield credibility on its own. That said, it could certainly wreak havoc with mass commenting campaigns and such.