> I could have Claude code shit out a... clone, and have none of the baggage...
Social media is different because of the network effects. Taking hold requires being there to catch refugees when something else collapses, and being sensitive to the fact of that collapse. It seems like quite a bit of luck is often required.
But you could also, you know, actually try to differentiate yourself. Figure out actual things that you'd want those sites to do differently, and have an opinion on which different approach to take.
And I especially wouldn't try with Twitter, because there are already at least two major competitors, and the people who have fled to them, in broad terms, seem to have done so based more on internal social strife than on any technical dissatisfaction. (Despite all the seemingly obvious technical issues to complain about!)
Please don't post like this. We could trivially get this take ourselves if we cared for it. It would be obvious that this is ChatGPT even without a disclaimer, and the analysis is exactly as formulaic and facile as you'd expect. (How could it reasonably conclude that pg's just-written essay is "controversial among economists and designers", let alone why? It's not making social media rounds; it was just published today; search engine results are mostly unrelated stuff and certainly aren't pointing to discussion...).
> Enough commenters didn't like the idea that everything they write publicly online is fed in as training data for AI that there's been a shift in this site's community.
Pardon; your theory is that this attitude was prevalent among people who like discussing pg's writing, and that they have left in favour of a new crowd that doesn't care about pg but is also pro- the AI companies?
... Because that doesn't seem to line up with the general tenor of discussion in threads about AI companies doing things.
> pg has the fortune or misfortune to write in a way that feels like some LLM writing, when he's writing well.
It pains me to think how simplistic some peoples' LLM writing detection heuristics are (or at least appear to be). Prose such as in TFA is really obviously human-written to me. It's using those choppy sentences properly. It doesn't strike me as "less well-written" at all; the resulting contrast is clearly very intentional.
Although, of course, what you describe is still a couple levels above "Behold, what doth mine Ctrl-F espy but U+2014 EM DASH! Hie thee hence, O wretched automaton!"
If you have showdead on, that user's comment history is rather full of this sort of thing. Seemingly restarted half a year ago, but with similar conduct in the pre-LLM era as well.
Okay, but how would it work? In particular, if AI is such a problem in the interview world now, how would it not be an even worse problem in the "taking tests for certifications" world?
Social media is different because of the network effects. Taking hold requires being there to catch refugees when something else collapses, and being sensitive to the fact of that collapse. It seems like quite a bit of luck is often required.
But you could also, you know, actually try to differentiate yourself. Figure out actual things that you'd want those sites to do differently, and have an opinion on which different approach to take.
And I especially wouldn't try with Twitter, because there are already at least two major competitors, and the people who have fled to them, in broad terms, seem to have done so based more on internal social strife than on any technical dissatisfaction. (Despite all the seemingly obvious technical issues to complain about!)
reply