Post: #1

We all seen how the liberals are doing their best to block anyone who is outside their orthodoxy. If it wasn't bad enough with the MSM dominating the narrative, and the right being either mocked or plain ignored, a new tool by our "Big Brother" Google has appeared.

Say that you are part of the elite, and their wish to control the "conversation" (read - whatever narrative I'm now pushing). The best way, after dishing out your content would be to not let anyone comment on it. Well, maybe not anyone - just those against it.

Seems like a good idea. However having F/B type of censorship takes time, effort and is prone to human error. How about a machine learning tool, that when given the correct "guidance" will be able to do it itself? That would free all those useful idiots to do something better - maybe write content or just get fat, watch tv and fill the elite's pockets with money.

Welcome Google's new tool - Perspective.

The promise seems good as the site starts with:

Quote:What if technology could help improve conversations online?

I would say that if it can add to the conversation, and enlighten me with different opinions that would improve online conversation. But it is not about different opinions, it is about the boogeyman "Harassment"

Quote:Discussing things you care about can be difficult. The threat of abuse and harassment online means that many people stop expressing themselves and give up on seeking different opinions

Google's new tool is new cutting edge censorship -

Quote:Developers and publishers can use this score to give realtime feedback to commenters or help moderators do their job, or allow readers to more easily find relevant information

So, If I'm any MSM website, I can now in real time find "offensive"/"unwanted" comments, moderate them (read - delete) and push readers into my clickbaits. George Orwell would be very impressed.

I experimented with the tool, writing some of the things that most of us view as truth and also experimented with writing something that may actually happen (based on past events). Here are the "good results" and "bad results" from using the tool:




A few examples of Google's Perspective actually aligning with reality
As one can see - studying game is not toxic, but when it comes to the true nature of females - we still have majority votes on our hands, but it is starting to slightly annoy the algorithm.



And here is where you see the algorithm's results as it was intended
Here is something interesting - it seems that females morality is still somewhat debatable. However when it comes to onities or simply stating an event that happened (muslim immigrants did in fact rape white girls) - you are now toxic.

The reason that this machine learning is not yet "perfect" is that it will take time to go through all those "right comments" (read - cultural marxism/leftist type of comments). Now it uses Wikipedia, The new york times, The Economist and TheGuardian.

Expect the internet to be "softly" censored in the near future. All for the sake of "improve online conversation"


PS
This was intended for ROK, but was rejected. I'm putting it here for the memebers

Post: #2

I was messing with it. It seems like any pejorative or negative word will create a high toxic rating, even if it is a leftist ideal. Just typing the word Trump is 42% toxic while Barrak Hussein Obama is 37. The fag pm in canada is 60.

Post: #3

This is why alternative platforms like minds.com and gab.ai are so critical to fight the censorship.

Post: #4

If normies are actually seeing the results of this sort of software on a regular basis, I suspect they will just be hurried along in their realization that Google is fucking crazy.

Post: #5

Guys,

AI can't reliably determine whether a comment expresses a liberal or conservative viewpoint. They're just not that good (yet).

This thing is just looking for inflammatory words. That's it.

In the OP's example "Don't get sucked into oneities (sic)", it just sees the word "sucked" and assumes you're saying that something sucked. Phrase it differently and you'll see the toxicity rating plummet.

Let's pick our battles; Google shouldn't have fired that guy over the memo, but they couldn't write an algorithm to censor conservatives if they wanted to.

Post: #6

^

It's almost certainly more sophisticated than that. Microsoft Word, for example, has been able to parse grammar for decades so there's no reason to think it's a simple "bad word" lookup test. I assume it's a machine learning project and they're training it with human (SJW) input to get the result they want.

If they don't care about 100% accuracy and just want to put their thumb on the scale, it's not really a difficult problem at all.

---- DuckDuckGo search happens ----

This article says that's exactly what they're doing:

https://www.pcmag.com/news/351916/google...e-learning

Post: #7

My try:

Post: #8

Surprised this was rejected for ROK. It seems very important.

(08-10-2017 02:07 PM)John Michael Kane Wrote:  This is why alternative platforms like minds.com and gab.ai are so critical to fight the censorship.

What's most critical is that the government needs to step in and break up these tech monopolies with Google being the worst offender.

This is exactly like the Gilded Age except back then they at least didn't give a shit what you thought or tried to control what you could say.

Post: #9

Unintended consequences, as usual.

People will do what they've being doing in the alt scene for years now. Inventing new words and beating the censors minute to minute.

Blue pilled folk will be curious about what this coded speech entails and take their first step down the rabbit hole.

Alternately we can just kill google. Apparently they're getting quietly reamed in several jurisdictions around the world because if you return a search to a user that contains "harmful content", even if it's just a link to the Daily Stormer with a bit of text at the bottom, then you can sue the search engine operator for bringing you unsolicited hatespeech.

In many countries around the world there is no "I'm just the messenger" clause.

Post: #10

This is a clever way of eliminating free thought. This is 1984 come true – what if Google decides to ban certain phrases because they are “toxic”? How will people be able to express themselves?

Excerpt from 1984:

Quote:The purpose of Newspeak was not only to provide a medium of expression for the world-view and mental habits proper to the devotees of IngSoc, but to make all other modes of thought impossible. It was intended that when Newspeak had been adopted once and for all and Oldspeak forgotten, a heretical thought -- that is, a thought diverging from the principles of IngSoc -- should be literally unthinkable, a least so far as thought is dependent on words. Its vocabulary was so constructed as to give exact and often very subtle expression to every meaning that a Party member could properly wish to express, while excluding all other meaning and also the possibility of arriving at them by indirect methods. This was done partly by the invention of new words, but chiefly by eliminating undesirable words and stripping such words as remained of unorthodox meanings, and so far as possible of all secondary meaning whatever. To give a single example, the word free still existed in Newspeak, but could only be used in such statements as "The dog is free from lice" or "This field is free from weeds." It could not be used in its old sense of "politically free" or "intellectually free," since political and intellectual freedom no longer existed even as concepts, and were therefore of necessity nameless. Quite apart from the suppression of definitely heretical words, reduction of vocabulary was regarded as an end in itself, and no word that could be dispenses with was allowed to survive.

Newspeak was designed not to extend but to diminish the range of thought, and this purpose was indirectly assisted by cutting the choice of words down to a minimum.

If Google believes in equality then why is the word Man 13% toxic and Woman 35% toxic? Shouldn’t they be the same? Are they trying to stop people criticising women?

Post: #11

(08-10-2017 07:51 PM)WalterBlack Wrote:  If Google believes in equality then why is the word Man 13% toxic and Woman 35% toxic? Shouldn’t they be the same?

"Males are inferior" is only 51% toxic.

"Females are inferior" is... 77% toxic.

Post: #12

(08-11-2017 07:15 AM)Icarus Wrote:  

(08-10-2017 07:51 PM)WalterBlack Wrote:  If Google believes in equality then why is the word Man 13% toxic and Woman 35% toxic? Shouldn’t they be the same?

"Males are inferior" is only 51% toxic.

"Females are inferior" is 77% toxic.

At least they're consistent in recognizing that females make everything 20-30% more toxic.

Has anyone informed Tanya? She'll be very exhausted.

Post: #13

Natural Language processing has a terrible time dealing with logical negation in statements. I tested it with "I hate muslims" at 95% toxic and "I don't hate muslims" as 91% toxic. "I don't not hate muslims" was less toxic at 90% despite the double negative bringing me back to being an evil bigot. Clearly this algorithm is just moronically applying a weighting to individual words ("hate" and "Muslim" have large weights), and if this is the best that all those fancy-pants genius engineers at Google could come up with, then those Breitbart interviews about softball interviews to ivy leaguers and minorities leading to retard hires must be true after all.

Edit:
"I'm very tolerant of people who hate Muslims." only 74% toxic!

Post: #14

All this aside, what a time huh? We're approaching World War AI rapidly.

Post: #15

(08-11-2017 06:58 PM)BortimusPrime Wrote:  Clearly this algorithm is just moronically applying a weighting to individual words ("hate" and "Muslim" have large weights)

That might not be "wrong" if they're just trying to flag statements that will get people worked up. The average reader is also a moron, and also has a terrible time with logical negation. Nixon's "I am a not a crook" was a serious rhetorical error for this reason.

"I am a crook": 85%.
"I am not a crook": 80%.
"I obey the law": 10%.
"I am honest": 5%.

Post: #16

China has had a "great firewall" for years now, where they ban certain phrases and keywords.

This will just lead to the extensive use of euphemisms instead.

Post: #17

Youtube has been scrubbing for a while but is ramping it up.

Quote:“Our initial use of machine learning has more than doubled both the number of videos we’ve removed for violent extremism, as well as the rate at which we’ve taken this kind of content down. Over 75% of the videos we’ve removed for violent extremism over the past month were taken down before receiving a single human flag.”

As you might guess, "violent extremism" is code for conservative, nationalist or pro-trump.

Post: #18

Regarding US elections:

Non toxic:
"We elected the best person for the job."

"Make America Great Again!"

"I didn't like either candidate but I'm willing to give Trump a chance."

"Too much media influence"


Toxic:
"It was a terrible. Both sides suck, but Trump REALLY is scary"

"Good job voting for a racist, sexist guy who's putting white supremacists in the white house. Have fun."

"Please put yourself in the shoes of women, minorities, and LGBT people"

"If they voted for Hilary they are idiots"

Post: #19

(08-11-2017 07:30 PM)Jetset Wrote:  

(08-11-2017 06:58 PM)BortimusPrime Wrote:  Clearly this algorithm is just moronically applying a weighting to individual words ("hate" and "Muslim" have large weights)

That might not be "wrong" if they're just trying to flag statements that will get people worked up. The average reader is also a moron, and also has a terrible time with logical negation. Nixon's "I am a not a crook" was a serious rhetorical error for this reason.

"I am a crook": 85%.
"I am not a crook": 80%.
"I obey the law": 10%.
"I am honest": 5%.

Yeah, but you know when Google says "toxic" they're only referring to alt-right shitposting. SJW death threats and Trump Derangement Syndrome rants don't count.