TikTok just added an AI off switch
TikTok has officially joined the “please no more AI slop” club. The platform has quietly rolled out a new slider in its Manage Topics panel that lets you tell the algorithm how much AI-generated content you want in your feed.
Yes. A literal slider for I’ve had enough of children riding alligators and Will Smith eating anything.
The new AI limit
TikTok first launched Manage Topics back in June as a way for users to say “nah” to content they’re not into. Now there’s a new option: Less AI-generated content.
In TikTok’s own words, it’s designed to “help people tailor the diverse range of content in their feed”. Translation: it won’t remove AI completely, but it should reduce the flood of glitchy fake faces, wonky physics and those bizarre AGT deepfake acts that look like Sims auditioning for Broadway.
But because TikTok can’t detect every AI clip yet, it’ll still sneak through.
TikTok is also testing invisible AI watermarks
TikTok’s next big move is invisible watermarking across content created using TikTok’s own AI tools (like AI Editor Pro) and anything uploaded with C2PA Content Credentials.
If you read our article on what C2PA is here to do, you’ll know that this is huge. It means TikTok will have better context on how a video was created, what was changed, and whether it was conjured by an AI model instead of a human with a phone and (sometimes) questionable idea.
But there’s one big gap: cross-posted videos. As soon as you save and reupload, a lot of metadata disappears. TikTok wants more platforms to adopt these watermarks to avoid the current game of “spot the AI”.
So what does this all mean?
Honestly? It’s a vibe shift.
Pinterest already added AI content limits. Now TikTok. Meanwhile, Meta is over there manifesting a future where your feed is 80% AI, your friends are mostly bots and you’re expected to… enjoy it?
If we spent that much on AI, we’d probably try to sell the dream too.
But here’s the reality: People don’t want feeds full of synthetic friends or perfect-but-empty visuals. Social media still works because it’s human. Messy. Weird. Unfiltered. We crave connection, not content engineered to death.
Right now, audiences are pushing back. And the platforms are listening.
But what about the future?
Gen Alpha will probably grow up treating AI content like tap water: just part of the infrastructure. Their threshold for “is this real?” will be wildly different.
But today, users want control, clarity, and frankly, they want their FYP to feel like their FYP again.
Expect other platforms to watch this closely.
TikTok isn’t stopping there
Alongside the new controls, TikTok is putting $2 million into funding creators who can teach AI literacy and safety on-platform. They’re also partnering with Partnership on AI to develop responsible AI practices.
We love a platform that realises you can’t just unleash the machine and hope for the best.
In closing
This could be a blip… or the start of a massive correction. Right now though, one thing’s clear.
People still want people. And platforms that put humans first will win.
This one's on us.
OOB is a social media agency. If this sparked something, let's talk.