Facebook building AI to recognize "offensive" memes

[dangerbunny17]

https://www.theverge.com/2018/9/11/17846512/facebook-rosetta-meme-machine-learning-text-recognition-hate-speech

 

"Facebook is building AI to help them out. In a blog post today, Facebook describes a system it’s built called Rosetta that uses machine learning to identify text in images and videos and then transcribe it into something that’s machine readable."

"Facebook says text extraction and machine learning are being put to use to “automatically identify content that violates our hate-speech policy” and that it’s doing so in multiple languages."

 

How will Facebook decide what is offensive and not offensive? What else will they be doing with this data?

[Tech_Dreamer]

[PlayStation 2]

For some reason all I can think of when Rosetta is mentioned is this https://en.wikipedia.org/wiki/Rosetta_(software)

[Wh0_Am_1]

11 minutes ago, dangerbunny17 said:

https://www.theverge.com/2018/9/11/17846512/facebook-rosetta-meme-machine-learning-text-recognition-hate-speech

 

"Facebook is building AI to help them out. In a blog post today, Facebook describes a system it’s built called Rosetta that uses machine learning to identify text in images and videos and then transcribe it into something that’s machine readable."

"Facebook says text extraction and machine learning are being put to use to “automatically identify content that violates our hate-speech policy” and that it’s doing so in multiple languages."

 

How will Facebook decide what is offensive and not offensive? What else will they be doing with this data?

Welcome to the LTT forums! Same way they do now with posts that they label "Hate speech" (almost anything can be found hateful by one group or another, so almost any conversation can be defined as hate speech, but generally they censor based on their political views)... Censor at their own discretion. 

[Zando_]

Imma say minion memes are offensive so Facebook will troll all the 40+ year old soccer moms for me. 

[matrix07012]

14 minutes ago, dangerbunny17 said:

How will Facebook decide what is offensive and not offensive?

*shuffles card deck*

[Andreas Lilja]

Kek... I mean lol.

[TacoSenpai]

But what if we just want to farm our memes and anime tiddies? 

[Syntaxvgm]

Hate memes. 

 

Less absurd than "hate facts" I guess. 

[Mihle]

Just ban all memes.

 

Problem solved.

[NumLock21]

People need to toughen up and stop being so sensitive.

[ScratchCat]

3 minutes ago, NumLock21 said:

People need to toughen up and stop being so sensitive.

That insults my inner idiot! _/s  People, if it is an offence report it otherwise ignore and get on with your lives

 

You can bet this is also being trained to detect anti-facebook memes.

[NumLock21]

5 minutes ago, ScratchCat said:

That insults my inner idiot! _/s  People, if it is an offence report it otherwise ignore and get on with your lives

 

You can bet this is also being trained to detect anti-facebook memes.

The problem is what may be a offense to you, might not be a offense to others. Like the Lyft taxi driver who got fired, because he had a Hawaiian hula girl on his dash board, and some SJW passenger got offended.

https://www.dailydot.com/irl/annaliese-neilsen-lyft/

https://knowyourmeme.com/memes/events/annaliese-nielsens-lyft-rant

[suicidalfranco]

Welp There goes all the digital black faces gifs

[W-L]

@dangerbunny17 The post just falls short of the requirements of the Tech News Section as of such it has been moved to General Discussion until updated. 

 

 

[Amazonsucks]

4 hours ago, dangerbunny17 said:

https://www.theverge.com/2018/9/11/17846512/facebook-rosetta-meme-machine-learning-text-recognition-hate-speech

 

"Facebook is building AI to help them out. In a blog post today, Facebook describes a system it’s built called Rosetta that uses machine learning to identify text in images and videos and then transcribe it into something that’s machine readable."

"Facebook says text extraction and machine learning are being put to use to “automatically identify content that violates our hate-speech policy” and that it’s doing so in multiple languages."

 

How will Facebook decide what is offensive and not offensive? What else will they be doing with this data?

Anything that doesnt go along with the idiotic ideology of their management will be promptly censored, just like James Damore and Google.

 

Silicon Valley CEOs are all of one mindset and anything that they disagree with(usually facts they find uncomfortable to hear in their echo chamber safe spaces) is considered "offensive".

 

I mean Bezos bought the Washington Compost just to push his agenda. Cuckerborg, Wojocki, Dorsey and the rest of those psychos are all pretty much identical.

 

Its no mystery that they are trying to censor specific people.

 

[Guest]

5 hours ago, Dan Castellaneta said:

For some reason all I can think of when Rosetta is mentioned is this https://en.wikipedia.org/wiki/Rosetta_(software)

AI has evolved. Rosetta is actually SkyNet, but more Apple-orientated. Can't wait till it gets control of the nuclear codes and create the iWorld.