How TikTok Is Rewriting the World


It is constantly learning from you and, over time, builds a presumably complex but opaque model of what you tend to watch, and shows you more of that, or things like that, or things related to that, or, honestly, who knows, but it seems to work. TikTok starts making assumptions the second you’ve opened the app, before you’ve really given it anything to work with. Imagine an Instagram centered entirely around its “Explore” tab, or a Twitter built around, I guess, trending topics or viral tweets, with “following” bolted onto the side.

Imagine a version of Facebook that was able to fill your feed before you’d friended a single person. That’s TikTok.

Its mode of creation is unusual, too. You can make stuff for your friends, or in response to your friends, sure. But users looking for something to post about are immediately recruited into group challenges, or hashtags, or shown popular songs. The bar is low. The stakes are low. Large audiences feel within reach, and smaller ones are easy to find, even if you’re just messing around.

On most social networks the first step to showing your content to a lot of people is grinding to build an audience, or having lots of friends, or being incredibly beautiful or wealthy or idle and willing to display that, or getting lucky or striking viral gold. TikTok instead encourages users to jump from audience to audience, trend to trend, creating something like simulated temporary friend groups, who get together to do friend-group things: to share an inside joke; to riff on a song; to talk idly and aimlessly about whatever is in front of you. Feedback is instant and frequently abundant; virality has a stiff tailwind. Stimulation is constant. There is an unmistakable sense that you’re using something that’s expanding in every direction. The pool of content is enormous. Most of it is meaningless. Some of it becomes popular, and some is great, and some gets to be both. As The Atlantic’s Taylor Lorenz put it, “Watching too many in a row can feel like you’re about to have a brain freeze. They’re incredibly addictive.”

In 1994, the artist and software developer Karl Sims demonstrated “virtual creatures” that moved in realistic ways discovered through “genetic algorithms.” These simulations, through trial and error, gradually arrived at some pre-existing shapes and movements: wriggling, slithering, dragging and walking.

But some early models, which emphasized the creatures’ ability to cover a certain distance as quickly as possible, resulted in the evolution of a very tall, rigid being that simply fell over. In doing so, it “moved” more quickly than a wriggling peer. It didn’t understand its evolutionary priority as “creature-like locomotion.” It needed to get to a certain place as efficiently as possible. And it did.

Swiftype Trends

Be the first to comment

Leave a Reply

Your email address will not be published.


*