3 ms·
I'm still not entirely sure how social is going to survive widespread adoption of modern LLMs. It was bad enough with the bots and bad actors we had in the twen
by MilStdJunkie 4y ago
I'm still not entirely sure how social is going to survive widespread adoption of modern LLMs. It was bad enough with the bots and bad actors we had in the twenty-teens, but it's going to be increasingly hard to justify ad dollars spent on a user base that's 99.9% ChatGPT screaming[1] at itself.
[1] "engaging"
- fullshark 4y agoWell 99% of what gets written by public figure social media accounts aren't actually written by the figure themselves but a social media/PR manager. And most people on social media aren't looking to talk to real people but just find bits of prose that confirm their biases 100% so, not sure it will matter. Though maybe the fog of "real or bot" just gets to be too much people lose interest.
- MilStdJunkie 4y agoThat's the thing - people will notice their biases being confirmed 100% of the time. All of their admired figures agree with them, and all the people they dislike are humiliated in just the right way, so that Everyone Knows I'm Right. There's going to be zero consistency. All of which, dammit, yeah, you're right, we have that already, but this doesn't have a leash on it in terms of human time. Like everyone having their own personal 24 hour news channel, each filled with gladhanding yesmen.