Unpopular opinion: bots might be a good thing for now.
I’m speaking from a growth perspective. Assuming users want to use social media to…socialize… you need active users and constant content. New social media platforms have a lack of users and content. Bots can bridge that gap until enough users are contributing and using the platform.
If you really think about it, it comes down to a platform using bots effectively. Let’s say the bots will only submit content when user submitted content falls below a threshold. Maybe it will auto generate threads for breaking news.
What if bots are used to ask questions and further conversations, like a social lubricant. Employed in a way to pull more useful information from users or to keep people engaged.
This all hinges on the ability for a bot to appear real.
This sounds super fucked when you think about it. I’m not a fan of bot content. If you didn’t know it was a bot, what difference would it make? LLM might be able to make it engaging and natural.
Imho that’s a horrible idea. A large part of content on the instance I’m on has become bots just reposting news articles without any own contribution, no discussion, nothing.
The go-to counterpoint being that people come to social media to socialize with other humans. The moment another “human” hits me with “As an AI…” or are otherwise unmasked for any reason is the exact moment I lose a little bit of faith in the platform.
It’s not enough faith to make me stop using it the first time or even the fifth, so long as the promise of almost always interacting with another person is dangled in front of me. But that little bit can’t be regained and eventually it’s going to hit zero and I will leave.
I already have chatbots if I want to talk to myself. Talking to the cat makes me feel less lonely than chatbots do, and given the choice between the fedi forever remaining niche or retaining the bot “activity” of reddit…I’d just move to tildes.
The only halfway good argument is the use of a breaking news bot, but I’ve found I tend to get tired of those very fast for the same reason. They just make me sad and irritated, and I end up blocking them. If the news is interesting enough, I expect humans will spam it.
If they could be programmed to only post when user interaction falls…maybe, in theory, but that feels more insidious to me than anything else. The idea of a company pumping their numbers will never make me like them, and if bots are already posting stuff, why do I have to interact in order to get content? They’re already doing it. 🤷♂️
If I’m lurking enough to trigger the theoretical user activity bot, I’d also be fine lurking while “other users” (the bots) give me things to look at, and they’ll never go dormant.
Unpopular opinion: bots might be a good thing for now.
I’m speaking from a growth perspective. Assuming users want to use social media to…socialize… you need active users and constant content. New social media platforms have a lack of users and content. Bots can bridge that gap until enough users are contributing and using the platform.
If you really think about it, it comes down to a platform using bots effectively. Let’s say the bots will only submit content when user submitted content falls below a threshold. Maybe it will auto generate threads for breaking news.
What if bots are used to ask questions and further conversations, like a social lubricant. Employed in a way to pull more useful information from users or to keep people engaged.
This all hinges on the ability for a bot to appear real.
This sounds super fucked when you think about it. I’m not a fan of bot content. If you didn’t know it was a bot, what difference would it make? LLM might be able to make it engaging and natural.
Imho that’s a horrible idea. A large part of content on the instance I’m on has become bots just reposting news articles without any own contribution, no discussion, nothing.
The go-to counterpoint being that people come to social media to socialize with other humans. The moment another “human” hits me with “As an AI…” or are otherwise unmasked for any reason is the exact moment I lose a little bit of faith in the platform.
It’s not enough faith to make me stop using it the first time or even the fifth, so long as the promise of almost always interacting with another person is dangled in front of me. But that little bit can’t be regained and eventually it’s going to hit zero and I will leave.
I already have chatbots if I want to talk to myself. Talking to the cat makes me feel less lonely than chatbots do, and given the choice between the fedi forever remaining niche or retaining the bot “activity” of reddit…I’d just move to tildes.
The only halfway good argument is the use of a breaking news bot, but I’ve found I tend to get tired of those very fast for the same reason. They just make me sad and irritated, and I end up blocking them. If the news is interesting enough, I expect humans will spam it.
If they could be programmed to only post when user interaction falls…maybe, in theory, but that feels more insidious to me than anything else. The idea of a company pumping their numbers will never make me like them, and if bots are already posting stuff, why do I have to interact in order to get content? They’re already doing it. 🤷♂️
If I’m lurking enough to trigger the theoretical user activity bot, I’d also be fine lurking while “other users” (the bots) give me things to look at, and they’ll never go dormant.
@Burp Agreed, bots might not be the best content, I’m not a fan, but if it drives engagement then does it really hurt at the moment?
@LambLeeg @Strangle