Slashdot Mirror
Zuckerberg Testimony: Facebook AI Will Curb Hate Speech In 5 To 10 Years (inverse.com)
An anonymous reader quotes a report from Inverse: After a question from Senator John Thune (R-SD) about why the public should believe that Facebook was earnestly working towards improving privacy, Zuckerberg essentially responded by saying that things are different now. Zuckerberg said that the platform is going through a "broad philosophical shift in how we approach our responsibility as a company." "We need to now take a more proactive view at policing the ecosystem," he said. In part, Zuckerberg was talking about hate speech and the various ways his platform has been used to seed misinformation. This prompted Thune to ask what steps Facebook was taking to improve its ability to define what is and what is not hate speech.
"Hate speech is one of the hardest," Zuckerberg said. "Determining if something is hate speech is very linguistically nuanced. You need to understand what is a slur and whether something is hateful, and not just in English..." Zuckerberg said that the company is increasingly developing AI tools to flag hate speech proactively, rather than relying on reactions from users and employees to flag offensive content. But according to the CEO, because flagging hate speech is so complex, he estimates it could take five to 10 years to create adequate A.I. "Today we're just not there on that," he said. For now, Zuckerberg said, it's still on users to flag offensive content. "We have people look at it, we have policies to try and make it as not subjective as possible, but until we get it more automated there is a higher error rate than I'm happy with," he said.
"Hate speech is one of the hardest," Zuckerberg said. "Determining if something is hate speech is very linguistically nuanced. You need to understand what is a slur and whether something is hateful, and not just in English..." Zuckerberg said that the company is increasingly developing AI tools to flag hate speech proactively, rather than relying on reactions from users and employees to flag offensive content. But according to the CEO, because flagging hate speech is so complex, he estimates it could take five to 10 years to create adequate A.I. "Today we're just not there on that," he said. For now, Zuckerberg said, it's still on users to flag offensive content. "We have people look at it, we have policies to try and make it as not subjective as possible, but until we get it more automated there is a higher error rate than I'm happy with," he said.
Why is there such an emphasis on this as opposed to any other type of "bad" speech? What makes hate speech inherently worse than offensive, but non-hateful speech, and how do you know if I hate you or not?
If humans can't even decide what is "hate speech", what makes anyone believe that an AI system can?
Correction: "Facebook AI will curb free speech"
Alive and well. Snowflakes have started started down a slippery fucking slope here. Life and society isn't all kumbaya, rainbows and unicorns. Pussies.
Oh look, a real world example of powerful entities planning to use AI systems for evil!
But go ahead, mock the AI skeptics with Skynet and Matrix strawmen.
We always seem to find ourselves in these little quandaries in our reckless pursuit of scientific development and apparently no one seems to think it's a problem. We can pass righteous judgement on the Soviets for their utterly insane development of nuclear technologies but then pat ourselves on the back for the latest social media spy tech disguised as a consumer product.
The Democrats and the SJW's have already decided what they consider hate speech.
Only the State obtains its revenue by coercion. - Murray Rothbard
What Zuckerberg will likely be "curbing" and what Facebook tends to ban:
- claims that racism is not the primary cause of poverty and criminality in minority communities
- pointing out that gender is not a social construct
- using a non-preferred pronoun with a transgendered person
- rude or critical statements about Hillary Clinton (if you use the words she/her, it's automatically misogynist hate speech)
- speech critical of illegal immigrants or advocating the expulsion of illegal immigrants
- speech critical or disapproving of Islam
- anything containing derogatory words for progressive protected classes (but not derogatory words for straight white males)
Of course, the net effect will be that Facebook turns even more into a progressive bubble. And while that may be comforting to progressives, it makes it hard for them to understand why their favorite political candidates or policies don't catch on among Americans in general.
They'll just follow the usual SJW rule which is "group X is incapable of $type_of_bigotry because they don't have 'power.'" See, if a black man grabs a white man and beats him to death screaming direct racial epithets along the way, that's not racist to them because the black man is a minority thus has no power, thus cannot be racist. Meanwhile, if a white guy from a trailer park, on SS disability screams "die n----" and shoots a super-rich black man that is totally racist because as a white man the trailer park denizen has white privilege.
This is why if actual Communists ever, God forbid, take power, I will have absolutely no sympathy for many of our kulaks (historically, "liberals" and "progressives" were about as hated by actual Communists as monarchists).
".. my conservative friends on Facebook.."
It is amazing what passes as a "conservative" these days.
It's even more problematic than that. If they have an AI algorithm which can understand human language they have the ability to filter out anything they do not like, not just hate speech and to misquote Agent Smith from the Matrix: "what good is the right to free speech if you are unable to speak?".
i know you wont, because it doesnt fit your worldview, but have a look at this:
http://www.ibtimes.com/white-black-crime-vs-black-white-crime-new-statistics-show-more-killings-between-2424598
Now, imagine what those stats look like if adjusted for total population..
Of course you are going to play silly qualitative games claiming 'but none of those black on white killings are because of race'
Believe what you will, however it is pretty common street knowledge that black gangs target whites because they are less likely
to resist violently, tend to carry more money, and deserve it more - and that is pretty much racism.
But no, your unverified single anecdote trumps decades of facts, doesnt it.
BTW, Yes, I know quite a few 'real' (maxist) communists, and even a real trotskiest.
The more serious 'believers' are very very much in touch with the fact that communism has a fatal flaw in implementation because it
centralises power excessively, and they know what that directly leads to.
As you obvious see Stalin and Mao as exceptions, perhaps you would like to point to the happy successful well balanced communist state?
The problem isnt left or right, it is totalitarianism, and anyone screaming for power for THEIR ONE TRUE BELIEF is pretty solidly in that camp from
my point of view.
Facebook, like Slashdot and Twitter, doesn't apply the same standard to everyone.
If you register a new account on Slashdot, your karma starts out low. You can post less often and don't get any karma bonus until you have built up your account a bit. Similarly on Twitter, brand new accounts that get immediately reported are often deleted without further warning, because they have a big problem with bots and people register new accounts to get around bans.
Facebook is similar. Once your account is established it gets more leeway. But once it has a number of "strikes" against it, posting something like a Pepe meme can get it banned rather than just warned. Pepe memes are mostly racist, particularly anti-Semitic, and associated with Nazism. The idea was to trick people by exchanging the easily recognizable swastikas for a cartoon frog, so Facebook tends to go easy on first time offenders, but after a pattern emerges they become less tolerant.
Note that I'm not defending Facebook, merely explaining this behaviour which seems to confuse a lot of people (like the infamous "I hate black people" tweet).
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC