This might make a dent in MKBHD’s problems
Yesterday, Google announced a handful of changes to YouTube meant to reduce issues with comment spam and impersonation. Among no longer being able to hide subscriber counts and a formal rollout of new content moderation tools, Google’s also trying to make it harder for channels to pick names that could be used to masquerade as others by limiting the “type and frequency” of certain special characters. This should make it easier to tell the difference between someone claiming to be the Android Police on YouTube and someone else saying they’re the 𝒶𝕟ⒹŘσ𝓘𝐃 𝓟Ｏ𝔩ᎥČε✅. All three changes might make a dent against the apparently rampant YouTube comment spam and scams.
The change was announced yesterday and is about as straightforward as described above. YouTube is imposing new restrictions on the characters available in generating a channel name going forward meant to reduce impersonation while also making it easier to see subscriber counts — together, commenters paying attention might more easily spot account “fakes” trying to impersonate well-known creators. YouTube’s new content moderation tools, in testing since at least April, can also increase how “strict” automatic filters for moderating comments are, reducing the chances that viewers will even see comment spam that leads them to scams.
In case you’re unfamiliar with the practice or why the character codes matter, there are a lot of Unicode characters out there which can resemble the plain alphabet many of us are used to writing in, and sometimes unsavory folk will attempt to impersonate a popular channel or individual by assigning themselves a name (and often an avatar) that can mislead the unwary into confusing an impersonator with the real person or channel. Throwing in checkmarks or other Unicode icons to look flashy also somehow lends an air of legitimacy in some peoples’ minds — perhaps they’re trained to recognize it through things like the Twitter verification checkmark.
It’s an issue on Twitter — you’ll often see fake celebrity accounts jumping in with cryptocurrency scams in replies to popular tweets — but also a problem for big creators on YouTube, who have to deal with impersonation from these other accounts in places like comments. It’s explicitly against YouTube’s policies but can be hard to programmatically detect. Popular Tech YouTuber Marques Brownlee/MKBHD touched on the issue a few months ago in a video.
You might think this is an easy issue to dodge, as Brownlee points out, but plenty of people fall for it, being led into scams at third-party venues like WhatsApp and Telegram. In many cases, those impacted blame content creators for the problem. Brownlee claims he wakes up each day to at least a dozen messages from those that are duped — people are a lot more gullible than you think.
Rather than trying to find new ways to detect it, YouTube has opted to limit “the type and/or frequency of special characters in channel names.” Based on the details provided, it sounds like this system hopefully shouldn’t impact other legitimate uses for special characters (channels serving content in other languages, as an example). Paired with improved content moderation tools like the increased strictness being tested and no longer being able to hid subscriber counts, YouTube’s changes may prevent people from falling for the sort of scams these impersonating accounts perpetuate.
The character restriction sounds like it will only be applied to channel name updates going forward, but the new comment moderation setting and disabling the hidden subscriber count are both being enforced already.
