From your comment's tone, it seems like this is supposed to be a bad thing. The only people who would be upset about this are folks who are trying to pass generated content off as real. I'm sorry if I don't have much sympathy for them.
Finding out a video is maliciously fabricated is a different problem.
Of course there was. The community guidelines have prohibited impersonation and misinformation for years.
That sounds like a quagmire of subjectivity to enforce. You can argue whether generated content was created to mislead or whether it would cause _egregious_ harm.
Now there's no more arguing. Is it generated or is it real - and is it marked if generated? I still fail to see the downside here.
> Now there's no more arguing. Is it generated or is it real - and is it marked if generated? I still fail to see the downside here.
How is this supposed to lead to less arguing? If there was an easy way to tell if something is AI-generated then you wouldn't need the user to tag it. When there isn't, now you have to argue about whether it is or not -- or if it obviously is, whether it then has to be tagged, because it obviously is and they've given that as an exception.
I'm not getting into the rabbit hole of finding out if it was actually generated - once again, that's a different problem. One that I already mentioned 2 comments back. My point was that there is less subjectivity with this rule. If the content is found to have been generated and isn't marked, then there are clear grounds to remove the video.
What youtube does when there is doubt is not known yet. I don't deal in "well this _could_ lead to this".
They use (with some irony) AI and other algorithms to determine if you're breaking the rules, often leading to arbitrary or nonsensical results, which the review process frequently fails to address.
> I'm not getting into the rabbit hole of finding out if it was actually generated - once again, that's a different problem.
It isn't a different problem, it's the problem. You want people to label things because otherwise you're not sure, but because of that problem exactly, you have no way of reliably or objectively enforcing the labeling requirement. And you specifically have no way to do it in the cases where it most matters because it's hard to tell.
At a mom and pop you could at least talk to a person and figure out what happen.
>I don't deal in "well this _could_ lead to this"
Did you not learn from the entire DMCA thing? Remember the thing where piles tech people warned "Wow, this is going to be used as a weapon to cause problems" and then it was used as a weapon to cause problems.
Well, welcome to the next weapon that is going to be used to cause problems.
There's a handful of cases where yt actually messed up w/ DMCA and considering the sheer volume of videos they process, I'd say it's actually pretty damn good.
So no, DMCA is not a valid reason to assume youtube will handle this improperly.
The alternative to the DMCA is Section 230 of the CDA, which could have just as easily been applied to copyright as it is to anything else if it weren't for the DMCA providing a more abuse-prone alternative.
> And people on the internet don't know what fair-use actually means, so they complain/exaggerate about DMCA takedowns when, surprise, it wasn't actually covered by fair-use.
Abusive takedowns are a huge problem, actually. It's common in cases of businesses sending takedowns for their competitors' websites or videos, for example. They're often completely fraudulent with no merit whatsoever, but the company receiving the takedown has no information on which to base a decision (who created this content? how would they know?), so they just mechanistically execute all of them with no validation.
The ban recourse problem is the opposite.
This is the "keep recourse": "this video is obviously bad, but google doesn't feel like taking it down, and there is nothing I can do about it". Now there is, and it can actually go to a court with a judge in the end, if Google is obstinate.
You didn't have a right to be hosted on Google before, and you don't have now. Of course they can ban you as they like. The thing is, they can't host you as they like, if you're breaking this rule.
Except that the rule can be satisfied just by labeling it, and if there are penalties for not labeling but no penalties for labeling then the obvious incentive is to stick the label on everything just in case, causing it to become meaningless.
To prevent that would require prohibiting the label from being applied to things that aren't AI-generated, which is impracticable because now you need 100% accuracy and there is no way to err on the side of caution, but nobody has 100% accuracy. So then the solution would be to actually make everything AI-generated, e.g. by systematically running it through some subtle AI filter, and then you can get back to labeling everything to avoid liability.
Did the California proposition 65 really result in cancer labels on everything? Or is it just hyperbole? I suppose having a lot of labels is still bad, even if they're not technically on everything.
Who's there
dmcAI takedown notice!
Really, your argument can be generalized to 'why have laws at all, because people will break them and lie about it'.
It isn't a rule against lying, it's a rule requiring lies to be labeled. From which you get nothing useful that you couldn't get from a rule against lying, because you'd need the same proof for either one.
Meanwhile it becomes a trap for the unwary because innocent people who don't understand the complicated labeling rules get stomped by the system without intending any malice.
> Really, your argument can be generalized to 'why have laws at all, because people will break them and lie about it'.
The generalization is that laws against not disclosing crimes are pointless because the penalty for the crime is already at least as severe as the penalty for not disclosing it and you'd need to prove the crime to prove the omission. This is, for example, why it makes sense to have a right against self-incrimination.
This divide is obviously going to play out on two sides.
Proving authenticity may turn out to be as difficult as proving fakeness. People will use this maliciously to flag and censor content they dislike.