Thanks, I hate it.
That’s the first thought that came to my mind when I saw this new feature in testing on Facebook, which enables you to create a post caption via generative AI, in order to help you come up with better content.
As you can see in this example, shared by Jonah Manzano, some Facebook users are now seeing a new “Write with AI” prompt in the post composer. Tap on it and Meta’s system will then generate a caption based on your existing post text and/or image, with cool suggestions like: “I mean seriously, what even is this day?!”. You can then prompt the tool to provide more options, including shorter text, a funnier description, one that’s more “heartfelt” etc.
So now, if you’re not funny or clever or witty in any way, you can get a bot to write your posts for you, and ride the collective output of the internet to newfound internet fame.
Which, in my view, runs counter to what social media is actually about.
Being “social”, by definition, involves human-to-human interaction, and the whole point of social media is to facilitate broader connection, not dilute it with robot updates.
Every generative AI addition erodes the core of social platforms a little bit more, and while you can also access such tools outside of social apps, so you can compose similar updates to these, even if they’re not made available in-stream, actually adding these options into the posting flow will undoubtedly increase their usage for social posts.
And really, I don’t even know if people actually want that.
I know that some people will use them, but the only real motivation for doing so would be to post better updates, in order to, what, get more likes?
So, let’s say, for example, that I take a viral post from someplace, I cut the text from that viral post and slap it into ChatGPT, then ask it to “give me a clever, witty response” to that message. I then post that response in reply. Maybe that generated reply gets a heap of likes as a result, but I’ll know that it wasn’t me that wrote it. It was a bot, and that bot is getting those likes, so what’s the point of actually making a bot provide this for you?
I can see why brands would do it, and scammy chumps. But for actual, real people who want to use social apps to connect with others, I don’t see why they’d want to use this type of tool to simulate a personality.
And if brands do use it (it’s not available for Pages as yet), their responses will also become increasingly distant, as businesses post AI-generated updates, then other users share AI-generated replies, while real people watch on from the sidelines, as social media engagement becomes a copy, of a copy, of a copy of the web that we once knew.
It just seems like the proliferation of generative AI tools is going to make actual human interactions less and less common, which will make all users more skeptical of every interaction, and thereby reduce social media engagement, at least in its intended sense.
But either way, every app seems to be jumping on the generative AI train.
It’s the tech trend of the moment, and as such, I get why platforms are looking for more ways to incorporate such into their streams. But I don’t think that direct posting on users’ behalf is the most beneficial use of generative AI tools.
And really, it seems like a road to nowhere for social apps.
But “engagement”, I guess.