Wikileaks: Why Hillary Clinton May Be Worried About Facebook's Artificial Intelligence

Mods - I only had a moment to do a quick search to see if this was posted somewhere else. IF so, please feel free to delete. Sorry for any inconvenience....

Wikileaks: Why Hillary Clinton May Be Worried About Facebook's Artificial Intelligence

Today, there was another interesting, yet underreported revelation from Wikileaks. In an email between Clinton herself and her campaign manager Robby Mook, Clinton appears concerned about Facebook's plan to open source their artifical intelligence. Specifically, she asks "Does this affect our plans?" So what does she mean?

I've read most of the emails that have been coming out for some time now. Lots of disturbing things in there...but this one keeps coming back to me. Why would Hillary care about Artificial Intelligence?

This is the article she was referencing -

Facebook Offers Artificial Intelligence Tech to Open Source Group

Facebook wants the world to see a lot more patterns and predictions.

The company said Friday that it was donating for public use several powerful tools for computers, including the means to go through huge amounts of data, looking for common elements of information. The products, used in a so-called neural network of machines, can speed pattern recognition by up to 23.5 times, Facebook said.

The first article I linked to goes through some speculation on what she might mean by "does it affect our plans?" I get a sinister vibe from her seemingly innocent comment. Maybe because it's coming from someone so corrupt.... I'm trying to separate out emotion and look at it from an objective, reasoning standpoint and yet, why would she care? Is it about the data collection and what it might tell people? I don't think so.... IF that were it, I don't think she would say "does it affect our plans?"

What are your thoughts? Has anyone else caught this?

a reply to: CIAGypsy

I would guess it is because AI operates on logic and without emotion. Finds patterns and such, so my guess would be that AI just might accidentally bring forth many things about how those in power have lied and manipulated the public to attain a certain outcome?


Feelings and lies do not equate to truth, where as logic and facts are a whole nother story........

a reply to: CIAGypsy

Depending on the date it may have to deal with facebook being more balanced as far as targeting certain demographics or excluding certain information...

a reply to: CIAGypsy

if artificial intelligence was capable of learning things, and not being biased by upbringing, location, surroundings, then it surely would realize the criminality the Clintons are all about and recommend they be executed.

at least that what I get out of all this.

originally posted by: CIAGypsy
Mods - I only had a moment to do a quick search to see if this was posted somewhere else. IF so, please feel free to delete. Sorry for any inconvenience....

Wikileaks: Why Hillary Clinton May Be Worried About Facebook's Artificial Intelligence

Today, there was another interesting, yet underreported revelation from Wikileaks. In an email between Clinton herself and her campaign manager Robby Mook, Clinton appears concerned about Facebook's plan to open source their artifical intelligence. Specifically, she asks "Does this affect our plans?" So what does she mean?

I've read most of the emails that have been coming out for some time now. Lots of disturbing things in there...but this one keeps coming back to me. Why would Hillary care about Artificial Intelligence?

This is the article she was referencing -

Facebook Offers Artificial Intelligence Tech to Open Source Group

Facebook wants the world to see a lot more patterns and predictions.

The company said Friday that it was donating for public use several powerful tools for computers, including the means to go through huge amounts of data, looking for common elements of information. The products, used in a so-called neural network of machines, can speed pattern recognition by up to 23.5 times, Facebook said.

The first article I linked to goes through some speculation on what she might mean by "does it affect our plans?" I get a sinister vibe from her seemingly innocent comment. Maybe because it's coming from someone so corrupt.... I'm trying to separate out emotion and look at it from an objective, reasoning standpoint and yet, why would she care? Is it about the data collection and what it might tell people? I don't think so.... IF that were it, I don't think she would say "does it affect our plans?"

What are your thoughts? Has anyone else caught this?

I want to start off by saying, great stuff here! Nice post

Now I want to follow up with my thoughts.

I think she has nothing to worry about, especially if she wins the election.

Artificial intelligence? As far as I'm concerned, the vast majority of those who frequent facebook are feigning intelligence anyway...I like to call it artificial, so if she wins, she has very little to worry about

I wonder if she was concerned about pattern analysis from this tool being used to highlight propaganda and where it was coming from (i.e. being planted)? They certainly seem to think it would be a bigger factor in 2020 election than in 2016.

Then there's the fact that millenials are more apt to receive their news and information from online sources, especially through Facebook, than through traditional media. A fact that I think the Hillary campaign neglected until it was too late...or at least didn't understand the potential of....

Still, I find it an ODD response to something that would seem so mundane. Given how corrupt she has proven to be, I wonder if this algorithm could be used to cheat?

originally posted by: CIAGypsyStill, I find it an ODD response to something that would seem so mundane. Given how corrupt she has proven to be, I wonder if this algorithm could be used to cheat?

Any algorithm can be used to cheat. The only algorithm that couldn't be is (ironically enough) an algorithm created BY Sentient Intellect.

An algorithm designed by man, can be manipulated by man.

My personal opinion, is she is trying to appeal to the ultra-conservative right. The bible belt. If she purports to being anti-AI, that would resonate with them.

a reply to: CIAGypsy

Neat post.

Initial thoughts is one of the following:

1. Hilary wondering if they can leverage the technology to get social media data for polling issues etc ...
2. Hilary wondering if it's possible to let the AI run the country given it mostly has the same skills as an American politician (pattern matching and regurgitating data)

a reply to: alphabetaone
excuse me, can you elaborate on that? what is your definition of "algorithm" and what do you mean with "An algorithm designed by man, can be manipulated by man."??

No clue what Hillary meant, but I do know why she'd be concerned about artificial intelligence. Pattern recognition is the means by which human intelligence and artificial intelligence put the pieces of a puzzle together. It's essentially what detectives do when solving crimes. You get all the data and look for patterns.
For Hillary, the problem is that for decades the Clintons have "debunked" accusations against them by attacking the motives of their accusers, NOT by addressing the factual information related to the accusations. With artificial intelligence, there is no "motive" --- patterns are recognized because those patterns exist, not because some biased person WANTS to THINK those patterns exist. The Clintons could not, for instance, brush off misconduct pointed to by artificial intelligence the way they brush off accusations by Judicial Watch --- calling it a right-wing organization that's been out to get them for years and WANTS to misinterpret information concerning the Clintons. The Clintons can't accuse artificial intelligence of being out to get them, and they have virtually ZERO experience in addressing FACTS in the absence of being able to deflect by launching their own accusations at their accuser(s). That's why the Clintons SHOULD be concerned about artificial intelligence. Whether Hillary is smart enough to realize that is a separate question. a reply to: CIAGypsy

originally posted by: Pinke
a reply to: CIAGypsy

Neat post.

Initial thoughts is one of the following:

1. Hilary wondering if they can leverage the technology to get social media data for polling issues etc ...

I thought about this too because it seems most obvious. But don't you agree that's an odd response (does this affect our plans?) to give if she can use it as leverage...

originally posted by: Pinke
a reply to: CIAGypsy

2. Hilary wondering if it's possible to let the AI run the country given it mostly has the same skills as an American politician (pattern matching and regurgitating data)

ha ha...star for originality here. I will agree that AI would probably do a better job of running the country than Hillary Clinton....

a reply to: WilliamBayer

Interesting perspective.... I may have to read it over several times to extrapolate some additional thoughts.... My initial thought is that your point is that they can use it to find patterns. Makes sense given that's what Facebook designed it to do.

However...

She said "does this affect our plans?" indicating a FUTURE action...not necessarily what it can provide from historical analysis. That's what is bugging me so much. What plans could they possibly be thinking of that they'd be concerned about pattern recognition and how it may affect them in the future?

Help me connect the dots....

originally posted by: verschickter
a reply to: alphabetaone
excuse me, can you elaborate on that? what is your definition of "algorithm" and what do you mean with "An algorithm designed by man, can be manipulated by man."??

An algorithm (because I design some contracturally) is a fancy way of saying a process or (in this case) a computerized method to attain a desired set of results, or a computation. Nothing more, nothing less.

I would think that the second part is pretty self-explanatory, but i'll follow-up none the less.... since humans (currently) design algorithms (for AI), certainly it is possible to design algorithms that either have backdoors attributed to them or are intentionally flawed to achieve results that simply skew from logic.

originally posted by: CIAGypsy
a reply to: WilliamBayer

Interesting perspective.... I may have to read it over several times to extrapolate some additional thoughts.... My initial thought is that your point is that they can use it to find patterns. Makes sense given that's what Facebook designed it to do.

However...

She said "does this affect our plans?" indicating a FUTURE action...not necessarily what it can provide from historical analysis. That's what is bugging me so much. What plans could they possibly be thinking of that they'd be concerned about pattern recognition and how it may affect them in the future?

Help me connect the dots....

Ok, i'll try and help you connect the dots.

There was chatter, amongst Clinton loyalists stemming from AI that suggested the FBI leaks were part of an Alt-Right's influence with their agenda being to discredit her.

I feel somewhat confident that (since the article you posted was on October 31st) she was worried if the AI code was released under the GNU GPL, that somehow their ability to show malintent by opposing factions could easily be rendered null or at the very least questioned to the point of the AI algorithm becoming an invalid argument. Though there are some who suggest she was worried about it far before January 2015....Since facebook tends to want to silence all but the HRC camp (it seems, I don't know about the reality of this) it seems facebook is her best friend right about now.

ATS thread

Maybe worried about a pattern being found in votes being cast on those machines that will prove the results aren't generated at random at all as they should be?

That is an interesting email - isn't it?

And Hillary - who is supposedly so technically deficient that she has to have her maid print her emails (classified or not) and writes in longhand at meetings rather than use a tablet or other device - is suddenly worried about AI?!

Odd doesn't even begin to describe it.

For what it's worth, I know a little about AI and the state of same throughout the industry. Facebook has some interesting stuff, but it pales in comparison to what Google has. Especially after their acquisition of DeepMind a British based AI company which was bought by Google in 2014.

I've seen others state that Google is in HRC's pocket, and if true, would explain why if she is worried about AI and her plans, she is not worried about Google's AI. Or perhaps, it may not be the AI tech itself per se, but rather the data mining one could do with it that has her worried about her "future plans".

Fascinating stuff...great catch!

originally posted by: Riffrafter
That is an interesting email - isn't it?

And Hillary - who is supposedly so technically deficient that she has to have her maid print her emails (classified or not) and writes in longhand at meetings rather than use a tablet or other device - is suddenly worried about AI?!

Odd doesn't even begin to describe it.

Actually, it's quite compatible with an uninformed technophobe.

There are a number of deep learning/neural network toolkits developed by big tech companies which are partially open sourced. Google has one, TensorFlow, and this is another, or perhaps a piece of software on top of a toolkit.

It is machine learning, not magic.

code.facebook.com...

research.facebook.com...


None of this is magic, it requires scientists to work at it, and lots and lots of clean training data. The AI will not kill you, it is a piece of software which, when commanded will produce some scores & numbers when you put into it an image, a data vector, or a sentence of words in a predefined format. A black box to crunch numbers.

As far as 'how does this influence our plans', all I can guess is that she would wonder if it would be useful for statistical election analytics, and direct mail marketing for political campaigns. And the answer to that is "probably not, more conventional statistical techniques are probably more appropriate for those problem domains."