I asked chat GPT to sharpen up my post A simple proposal for preserving free speech on twitter. It responded with a much shorter post. I decided that I preferred my longer exposition, so kept it, but I'm interested in an unbiased opinion of whether I made a mistake, and a shorter, snappier, but less detailed post would have been better so that I know for the future.

Here's the rewrite:

Twitter faces a constant challenge in balancing the principle of free speech with the rights of individuals to not be subjected to speech they find unpleasant or distracting. In everyday life, people can express their opinions freely, and those who do not wish to hear them can simply avoid those places or spaces where they are being expressed. However, on Twitter, this approach is not as effective. One person can reply to thousands of people who have not expressed any interest in their views, and their responses may be seen by millions. The current approach of banning users from Twitter can significantly impede their ability to express themselves, especially considering Twitter's large user base.

A simple solution to this issue could be to prevent users who would otherwise be banned from replying to non-followers. This would allow them to express their opinions freely, but would prevent them from forcibly communicating with people who are not interested. This approach would also allow for more popular figures, like Donald Trump, to continue to receive engagement on Twitter, while those with less interest in their views would be unable to reach a wider audience. This approach would preserve free speech while also protecting the rights of individuals to avoid unwanted speech.

New to LessWrong?

New Answer
New Comment

1 Answers sorted by

I didn't particularly like either one.  GPT starts off with clarity that it is Twitter's challenge, not a general discussion, which I'm not sure was your point.  Actually, I'm not sure what your purpose was in either case, so that probably just follows.

Neither one acknowledges the weakness of the argument - it's simply false that most speech is permitted in almost any large private forum.  There's both explicitly forbidden speech, and speech that is socially punished (by other users). Also, most venues are not take-it-or-leave it at a fine-grained level - you can leave entirely (as you can Twitter), or you can accept their bundle of speech that the venue and community allows).

4 comments, sorted by Click to highlight new comments since: Today at 4:59 PM

I do find this one generally better, since I think the overall discussion is not that novel and so doesn't need such long explanation. That said, it did lose a key justification from the other post, that Twitter's size makes it akin to the public square, which is why this kind of rule is necessary to contemplate.

What were your prompts?

Can you sharpen up the following blog post:

I noticed that the original version chat-gpt suggested made it sound like all users would be banned from replying to non-followers, not just those who would be banned. I told it to clarify, and it did, but that highlights one of the risks of relying on chat-gpt - you have to proof read very carefully.