TheMaster666
Potential
- Joined
- Aug 17, 2021
- Messages
- 320
Has anyone noticed Youtube bizarrely randomly auto-deleting comments moments after posting them? I've noticed this for a few days. It looks like the comment went through, but if you reload the page or sort by "newest first" then back to "top comments" it magically disappears. This doesn't happen all the time, but seemingly half the time.
I was watching The Onion, and they had a great video which was like a video equivalent of a political cartoon on gun violence in America. A man is pointing a pistol at his head, so he responsibly decided to buy a rifle to protect himself from himself.
Chuckling, I just said this is a good metaphor for the gun violence debate. Banned.
Okay, I thought, maybe violence triggered the algorithm? Tried again rewording violence to gun control debate.
Deleted again.
This is very strange and more than a little Orwellian the way major private companies have such powers of censorship. I understand free speech doesn't apply to private organizations who are free to set their own rules as they see fit, but that policy only makes sense for smaller scale sites like this one, for instance. For the massive social media organizations that everybody uses, is corporate censorship substantially different from state-based?
Given the scope and volume of comments out there, no moderator-based approach would be feasible (and it's not like Reddit moderators have a particularly good track record either anyway on the matter) so a.i. - based filters that auto-target certain key phrases seems most practical to implement, but the false positive rate seems to be really bad, i.e. picking up non-issues as though they are issues.
Anybody else noticed this lately?
Edit: Looks like a few minutes later I can see my comment now on this particular video, but I've still noticed this phenomenon on other comments.
I was watching The Onion, and they had a great video which was like a video equivalent of a political cartoon on gun violence in America. A man is pointing a pistol at his head, so he responsibly decided to buy a rifle to protect himself from himself.
Chuckling, I just said this is a good metaphor for the gun violence debate. Banned.
Okay, I thought, maybe violence triggered the algorithm? Tried again rewording violence to gun control debate.
Deleted again.
This is very strange and more than a little Orwellian the way major private companies have such powers of censorship. I understand free speech doesn't apply to private organizations who are free to set their own rules as they see fit, but that policy only makes sense for smaller scale sites like this one, for instance. For the massive social media organizations that everybody uses, is corporate censorship substantially different from state-based?
Given the scope and volume of comments out there, no moderator-based approach would be feasible (and it's not like Reddit moderators have a particularly good track record either anyway on the matter) so a.i. - based filters that auto-target certain key phrases seems most practical to implement, but the false positive rate seems to be really bad, i.e. picking up non-issues as though they are issues.
Anybody else noticed this lately?
Edit: Looks like a few minutes later I can see my comment now on this particular video, but I've still noticed this phenomenon on other comments.
