Hacker News Reader

Comment by monocasa | Hacker News Reader

It doesn't. LGBTQ is a category that gets special cased in order to block homophobic creators. But it mainly just blocks pro-LGBTQ creators because the homophobic creators are more likely to speak in dog whistles that the blocking algorithms don't understand, whereas the pro-LGBTQ creators speak frankly and quote homophobic speech in order to comment on it.

View on HN

A lot of these suers are just making videos of sex toys[1] and other stuff that is very well known to be sensitive stuff despite the irony of all pop culture being about sex. This lawsuit has nothing to stand on, even people making videos of electronics get striked, harassed, deplatformed. They just assume it's because of lgbt, but it's standard YouTube.

https://youtu.be/0rjrA2GJ1lo

LGBT creators are not a special case, all news and politics commentators I follow are demonetized, all gun related channels as well.

That's probably a sign that you're being radicalised and should find some saner news and politics commentators.

Do you apply the same logic to LGBT content that is complaining about getting demonetized?

They're demonetized all across the political spectrum. Be it right wing like Crowder or left-wing like Tim Pool.

Doesn't sound like Tim Pool is really all that left wing: https://rationalwiki.org/wiki/Tim_Pool#Clickbait_content

That is a rather disgusting accusation to make, and one that eimultaneously indicates a political bias and presumes that automated moderation decisions are always correct.

If this is the case, that's tragic in the dramatic sense -- LGBT groups fighting algorithms that are protecting them, because they can't appreciate the unseen danger prevented.

I also imagine the 'algorithms' are a moving target, which makes me think this should be settled quickly if not dismissed, as Google tries to correct this. But certainly a lawsuit should bump this up the engineering priority list.

I hope that is a the case of it being sorted quickly though I am unsure they can teach intent well enough to actually make this any better.

> If this is the case, that's tragic in the dramatic sense -- LGBT groups fighting algorithms that are protecting them, because they can't appreciate the unseen danger prevented.

I don't think the article or GP are saying this. The algorithms, while ostensibly well-intentioned, are actually having the opposite effect — stifling the speech of the LGBTQ+ community while allowing their harassers to operate mostly unchecked.

If they block 40% of real hate speech but also 5% of false positives (lgbt commenting on hate speech) I would say the algorithms are working overall.

I have no way to know the real numbers but I am assuming Google monitors overall effectiveness of algorithms and only use what works.

If they're having trouble classifying, they might not have good data on effectiveness.

You would have human review of a random sample to test effectiveness. In fact, this is usually how the ML algorithms are constructed in the first place, e.g. supervised learning. So yes I think they should have this.

Okay, but why do you think the numbers look anything like that? Conversely, if they block 5% of hate speech but are 40% false positives, then the algorithms are clearly not working. We have no reason to assume they're effective, and in fact this lawsuit contends that they're actually harmful.

Well, the algorithms simply don't work if they're flagging too many false positives.

Reply on news.ycombinator.com