It's hypertailored compared to YouTube and the likes. When people complain about the content they see, be it terrible DIY hacks or something more bizarre, I can't help but wonder what they've done that TikTok recommends such content to them. After dropping a few hints that it's very tailored a few have gotten a bit embarrassed and stopped complaining.
Interestingly it also separates a lot of the users from the rest, people call them [thing]-toks and crossovers aren't very welcome. Kinda like subreddits, but algorithmic.
This has been my experience--once and awhile a new topic will start showing up in my feed, say 1%'r motorcycle clubs or peanut-butter-whisky drinkers. All it takes is a couple "not interested" clicks and poof, back to people showing off vintage computers and HAM radio videos.
> To that end, sometimes you may come across a video in your feed that doesn't appear to be relevant to your expressed interests or have amassed a huge number of likes.
> By offering different videos from time to time, the system is also able to get a better sense of what's popular among a wider range of audiences to help provide other TikTok users a great experience, too. Our goal is to find balance between suggesting content that's relevant to you while also helping you find content and creators that encourage you to explore experiences you might not otherwise see.
The underlying narrative here that most people do not grasp is that each user's experience is different than ours, and their preferences are different too.
If someone else is complaining about their experience, we shouldn't be dismissive of it, it could easily be a real concern based on dark patterns the platform leverages.
For example, the platform can amplify triggering content to someone who exhibits schizophrenic traits, or it can show self harming or depressing content to someone who exhibits bipolar traits.
It's harmful to assume algorithms can't be implemented without flaws, agenda, and human bias... Actually it's willfully ignorant to think that anything inherently made by humans can be made without flaws and bias.
Interestingly it also separates a lot of the users from the rest, people call them [thing]-toks and crossovers aren't very welcome. Kinda like subreddits, but algorithmic.