What Social Networks do with your data (Part 3)
I've been a fan of Project Veritas for quite some time, but now they have gone to the next level and exposed how Twitter abuses the data people entrust to it. In this blog I will break down what Twitter and most certainly other big data "Krakens" like Facebook, Google (and its parent company Alphabet), Microsoft, Apple and many more do with the data users grant them. This episode focuses on the first of the undercover videos that Project Veritas has filmed of Twitter employees telling in what they would never want their users to hear.
Note that the actual video starts at about 01:32. You can find it here
Picture sources: here and here
Embedded (Please go to 01:32):
What Social Networks do with your data (Part 3)
Social Networks have no decency, recognition of the law or privacy protection. Deal with it. You can't even get away from them if you want to
The video begins with Pranay Singh, Direct Messaging Engineer of Twitter, introducing himself and explaining what he does. When asked if he could hack someone's DMs (Direct Messages), he evades the question, though we already know from Clay Haynes in part 1 (here) that they can:
We can absolutely look at every single message, every single tweet, whatever you log into [...]
After some seemingly drunk talk again stating the possibility and a bit of information from Project Veritas, he continues:
So, everything you send is stored on my server...
IMPORTANT: He talks about HIS server, not a Twitter server. This means that he in essence makes a copy of all Twitter tweets and messages, possibly more, on a private server. It's even possible that all the users send is truly copied, though that would take up a lot of space. Nonetheless, if he copies all data, this raises a question: What if someone hacks this server? Would they have access to every DM or Tweet ever written, or at least going back to when he first created this system? This is a serious security issue. He continues after being asked if one can delete it:
You can't, it's already on my server now...
Not that it would make any difference to Twitter in general. We know, again from Clay Haynes in part 1:
We have full access to every single person's account, every single direct message, deleted direct messages, deleted tweets.
But now we know there are at least two copies of if out there, not including backups that Twitter surely has. Singh continues:
So all your sex messages and your, like, dick pics are on my server now...
Not only does he store textual information, he also keeps pictures. I can only imagine how much data storage this must cost, but I guess having this data available does create the possibility of blackmail. He continues:
All your illegitimate wives and, like, all the girls you've been f**king around with, they're on my server now...
and
I'm going to send it to your wive, she's going to use it in your divorce.
In other words: He has the ability to blackmail people whenever he wants. With his job, he could probably contact the individuals on Twitter and get away with it without any trace. But note that, as there is currently no evidence of such action, I am not accusing him of it. Continuing:
So, what happens is, like, when you like, write stuff or when you post pictures online, they never go away. Like, they're always on there.
Going back to Clay Haynes, that's old news by now. Next he talks about how Social Networks and just about any website that employs ads and tracking or fingerprinting software makes money:
Uhm, because, like, even after you send them, people are, like, analyzing them, to see what you're interested in, to see what you're talking about, and then sell the data.
It is interesting that he talks about people analyzing the data when we already know from part 2 (here) that Twitter employs AI technology to analyze the content that people produce and throw out what they don't like. It seems that this technology is not yet fully ready and requires human oversight, or was never intended to work fully independent in the first place. But other than that, it is extremely common, and by that I mean almost 100% of all websites that people frequently visit track the users through tracking, fingerprinting or cookies and use the data themselves and/or sell it, receiving specifically tailored ads for every person visiting. Of cause Twitter with its access to conversations between people would employ this too. May even go as far as to create a virtual profile for people who aren't on Twitter or have interacted with other sources, but are mentioned by others. Imagine it like this: Your daughter is 5 years old and has no Twitter and you tweet to your wife that she would like a doll for Christmas, then when she's older and gets a Twitter account, Twitter recognizes her and goes off the data you sent before, sending her the most recent Barbie doll ads. Singh himself confirms that this data goes to advertisers when asked. He then goes on to confirm that everything goes through this data collection process and then goes on to confirm once more that machine learning is employed for this process:
A machine is going to look at it. An algorithm will look at it and they'll make a virtual profile about you.
The video then switches to Mihai Florea, Software Engineer of Twitter who specifies:
To actually charge advertisers for money, we have to prove it was you, and that's why we're using, like, the email address, or like, a cookie, something that can track you.
Again, the fact that advertisers require such proof only shows how ubiquitous this has become. One may very well assume that everyone does this on every website at any given time. The moment you click the "like" button of a Social Network a script is set into motion that adds this bit of data to the already impressive collection. The same happens when you click onto a link or, especially, and ad. And it seems to pay well, looking at the stocks of both advertisers and Social Networks, combined with Florea confirming:
That's how we make most of our money.
In part 2 I've talked about the very likely possibility that Twitter is also charging people in both high power and with a lot of money if they want others banned. Again, I might be wrong on this one, but it's a possibility and it won't fail on the available methods nor the motivations of either side of the deal. Florea continues when asked about "signing away privacy" at Twitter:
Yeah, I mean, yeah. You're paying for the right to use our website with your data basically.
I have no idea where this originated from, but please remember: If something is free, YOU are the product to be sold. He then goes on:
And it's the same with every free website.
And where it's possible, I'd assume every paid website too. Greed has no threshold that when reached satisfies, it only shows the next milestone to reach. And then the next, and the next, and so on. Next, the video goes to Conrado Miranda, Former Engineer of Twitter. He elaborates:
You leak way more information than you think.
and
Like, we have information from people... Like, if you go to Twitter for the first time, we have information about you.
Exactly as I told you before, but now we're at the part where it's confirmed. And the next part if you read to this point you'll know too:
Because, like, we actually bought a company because of this, like, ads network. So like, when you... We serve ads to other companies, in other people's sites.
The "free service" online market is not free. It charges you with data and takes it without you even realizing. Of cause people could read tons of pages of agreements and terms of service, but who has the time for that? In essence, replace them with "We'll take every last bit of data we can mine from you, and we'll use it however we can get the most money out of it because we are greedy pieces of sh*t and your privacy doesn't matter a drop of water less in the ocean!", that would be all in all libraries worth of lawyer-speak summed up in one or two lines and could be applied to just about any website out there. And Miranda confirms this:
So, when you go in and see that ad, we get information that you're this person, seeing this ad, at this moment at this location on this website.
and
So, we already have a lot of information about you.
When asked what would happen if it fell into the wrong hands and how he'd protect people from that, he answers:
You don't.
and
There is no way.
The most honest answer would have been "you couldn't even if you try", neither stopping others from getting this data if a Software Engineer can just collect data freely as he pleases nor do anything about it because while Twitter could in theory delete anything that they have, others won't throw away data just because Twitter says so and of cause they do have everything backed up themselves as well. The video then continues to the middle man himself, Clay Haynes, who states:
It is creepy Big Brother.
From part 1 we already know that he is kind of scared himself when he states:
It's very, very dangerous. Also, very, very creepy Big Brother-ish.
But as he sees it either as a job he gets paid for or from his self-proclaimed "bleeding-heart liberal" personality as a way to advance his agenda, he is still doing it. And once more, he goes:
I mean, it's like a level... I don't want to say it freaks me out, but it disturbs me.
He continues after that:
I get to look at all of the reported tweets, which means I've seen a lot of dick pics. It's ridiculous.
So remember: If you are female and pay too much for all your dirty websites (that by the way 110% track you by all possible means too), then apply to Twitter and presumably Facebook and others. You'll have a lot of fun and can mess around with people's private lives as long as you don't anger your superiors. The conversation continues with the journalist asking how these are kept from popping up. Haynes answers:
There are bots. There's something called machine learning, which basically, when someone reports that picture, it gets put into a database and then as more of those pictures come into place, the machine... our services learn...
A basic explanation of current AI technology: You give the robot a database of related things, it looks at the data and finds similarities (in the case of "dick pics" probably a centered longer piece of flesh-colored pixels), then it can find similar things in a database that has both what the algorithm is supposed to find and supposed to discard. The journalist goes on stating that this is not appropriate, Haynes responds:
Yeah. And also, more along the lines of they can actually start guessing, like, is this... is this equipment or is this maybe, like, an elbow or something.
Probably until the point where the algorithm gets so much data from you that they know the last 10 people you slept with from that picture even without having that data from other sources. But all jokes aside, it really is the biggest collection of porn out there: The images and videos people uploaded and of cause the entire NSA and CIA espionage through backdoors, accessing your smartphone camera whenever they want to. If you want to know the extend of criminal material, just imagine how many little children are playing angry birds, candy crush & co while sitting on the toilet taking a dump and pinpoint where the camera's field of view points the whole time. This leads me to believe that the biggest collection of child porn to ever exist lies on the databases of US and Chinese intelligence agencies. But on with the video. Brought up by the journalist's reference to a former conversation, likely from part 1 of this series, he once more states that nothing is ever deleted. Later on, he continues:
I've seen way more penises that I've ever wanted to see in my life.
I'm just betting there's leftists out there crying "homophobia!!!" when they hear this. But on with it: He is asked if this is from direct messages and confirms. Later O'Keefe states it sounds like there's a lot of porn involved in what he does. Haynes answers:
Unfortunately, there is a lot of porn.
and more:
That's, yeah... You know, actually... This sounds horrible, but I'm actually glad and fortunate it's just dicks, it's just blow job pictures, it's just that type of stuff.
Which makes me wonder what the bottom of that rabbit hole looks like. I can only imagine what the people who have to review things for removal get to see, from [child] porn and rape to murder and really ugly accidents. 99,9% of people would probably never upload anything more disturbing than what they did last night and with who, but there are really ugly things out there. This is a job for people who have a strong stomach or are just devoid of any emotions, and I do not envy them. But of cause, as we know from the very beginning of this, some people seem to store, view and use it however they see fit. O'Keefe states that this is scary to him, Haynes confirms. He is asked again if he has access to all DMs and confirms once more. He is then put onto the subject of such data leaking:
Oh yeah, and it's a genie out of the bottle kind of thing after that point. You know? Sure, I can fire them. Heck, I could probably even sue them, in some cases.
and
But the genie's already out of the bottle. Like, how do actually recoup costs... You can't calculate the cost or the damage of that.
Just for a moment imagine the thunderstorm of lawsuits of people whose privacy has been violated by Social Networks rolling to the HQs of Facebook, Twitter and Alphabet (Google's parent company), only to be shut down by pages upon pages of lawyer-speak in user agreements and terms of service. This is their imagination of "privacy": You don't have it, because we want money and power. The video continues with more audio:
There's teams dedicated to it. I mean, we're talking, we're talking three or four... at least, three or four hundred people.
Likely the people who, in all fairness, have to exist and look at these pictures in order to delete what people really shouldn't see. As I assume Twitter has to employ many more than just three or four hundred people (significantly more) to look at reported content, it's possible that they have used algorithms to pre-filter pictures and certain parts of the review agents have certain subjects they are tending to. But also they could simply insert an USB drive and copy whatever they want. There are methods to track and prevent this, but no software is without bugs or backdoors, and of cause if you are literally at the source [code] like Software Engineers are, anything goes. And it seems the people who are literally paid to do this, as Haynes states, have a lot of stuff to look at:
You know, there's girls with full spread-eagle pictures on there.
And later he confirms that even Pornstars do this (though, let's be honest: That doesn't shock anyone anymore). According to Haynes they do it, in short, for followers.
The video, and with it the series, finishes off with a look to Facebook and Google. A journalist asks Florea presumably about tracking in relation to other companies, and he answers:
We know some stuff. We aren't as creepy as Facebook or Google, but we know some stuff.
So then let's ask ourselves the final question: How in the world can one avoid being spied on? So here are my 5 golden rules for avoiding this sort of stuff:
1.) Don't get your private information on Social Networks. Once the genie is out of the bottle there's reversing it, but if that doesn't happen no need to worry.
2.) Don't get private information of anyone else on Social Networks. As we've seen, they will use whatever means they have to get as much data as possible.
3.) Disable your device's cameras and microphones. The best would be a small noise-canceling pocket to put it into. It should also have metal woven into itself to stop signals from getting out and revealing your location and block out the radiation from damaging your cells. Though as most people won't do this, or so I would imagine, Big Data will likely know all the time where you are and what you do anyway.
4.) Avoid browsing on websites without protection: Have a browser that blocks ads, trackers, fingerprinting software, browser scripts and cookies is helpful, having a VPN will also hide your true location. Of cause, most websites won't be able to deliver the services they offer without all of this or tie their services to such data collection methods. So whenever you do allow any of this, be aware that you are putting out information on yourself.
5.) The most important one: Share the information of what you've learned with others and show them the importance of their data and the possible consequences of ignoring its influence. Just last week one of my teachers told a story about a family member (if I remember correctly) seeing one of her vacation pictures that she never shared but only kept stored for a while and deleted afterwards on an ad. It is just as easy to put everyone sexting and sending "dick pics" onto any website that would be willing to pay for it.