[chologrande]

Lots of comments here chiding OP for running or talking about this; I think there's something to learn here.

This is just a hobbyist prompt and api. If nothing else, I'd say at minimum this highlights that there are likely much larger farms that have been operating in a similar way but at larger scale for longer periods of time (but not talking about it)

[M4v3R]

Exactly this. If one person with spare time on their hands can do that as a side project, one can only imagine what a government-backed actor can do with this technology. I would not be surprised if in a few years we suddenly learned that half of internet comments are bots, similarly to what happened to email (most of email traffic nowadays is automated emails & spam).

[theodric]

And after all, whomst among us isn't really just three government AIs in a trenchcoat? I know I sure am.

[Litost]

Is this what happens nowadays when you sign up to a dating app? I mention this only slightly tongue in cheek:

a) Because it is being argued by some as a good thing that you have an AI do your flirting for you? "we can free people from writing a thousand introductory messages, giving them energy to focus on the humans on the other side." https://www.theguardian.com/commentisfree/2023/jun/06/ai-bot...

b) As someone mentioned the turing test, as a comment on the Ashley Madison scandal someone did suggest the chatbots on there had passed it. "Claire Brownell suggested that the Turing test could possibly be passed by the women-imitating chatbots that fooled millions of men into buying special accounts" https://en.wikipedia.org/wiki/Ashley_Madison_data_breach

[JohnFen]

So we can add the destruction of the web to the risks AI poses? Is there nothing AI can't lay to waste?

[Levitz]

It really does look like dead internet theory becoming real is a matter of time.

[hammyhavoc]

What is there to learn? We already know that it's possible. This is some of the lowest-hanging fruit imaginable and is what spammers have already been doing.