So my friend was telling me yesterday that she got a new Samsung smart TV for her living room. One afternoon she was sitting on the couch across on the TV, which was turned off, talking to her friend about social anxiety.
The next day her Samsung phone, Facebook account, and her son’s Instagram account (which is linked to her phone number) were blown up with ads for social anxiety medications.
She called Samsung up like “WTF“ and they admitted that it was a “feature“ on the smart TV for it to record conversations even when turned off in order to provide customized ads. You could turn off the setting but of course you probably wouldn’t even know about it unless something like this happened because it’s not like they readily advertise “this thing will record your conversations even when it’s turned off“.
this ain’t alright with me … bitch what the fuck
https://www.cnet.com/news/samsungs-warning-our-smart-tvs-record-your-living-room-chatter/
This is so much worse than Mark Zuckerberg. This is literally some 1984 shit. This will make it so easy to spot people who don’t, umm, agree with the government, and in a fascist state that would be a great surveillance method. Since America is nearing that with Trump as president,
TURN OFF THIS FEATURE ON YOUR SMART TVS Y’ALL
Hot take: don’t buy them in the first place
Being the last person in 2030 not to have any smart devices in your home will probably make you a suspect.
Just like turning off multiple smartphones to have secret conversations can already make you suspect NOW.
Organize against listening features and against State power to use them, but adapt your own security culture in the mean time so your data use looks ‘normal’.
Guys, there are WAY too many people for our existing government to watch everyone! It would be like finding a needle in a haystack, and even then this shit happens SO much that false positives are so common that the true positives probably got missed a long time ago. Honestly, the NSA paranoia that came up is simply so the government could lie to a select few individuals that they were watching and keeping security.
So, this is gonna be a loooooooong response and you don’t need to read it if you’re not interested. I’ve been meaning to write a post about this so I guess the comments on this post make a good draft space…
You are right, there are far to many people for a government to effectively listen to all at the same time. BUT tthe availability of a massive amount of data + very fast analysis software means that they no longer have to do that. I’m gonna take a moment to explain how intelligence agencies can use data analysis to decide who to watch, and give some clues of how you (’you’ being anyone reading this) can stay ahead of their game.
Wanna join me? It hinges on two parts: Networks & Patterns.
Networks: this one is easiest to understand. You have online networks. Your facebook friends, the people in your phone adress book, the dropbox you share with coworkers, everyone who bought online tickets to the same punk band you attended. Take your networks, combine them with other people’s networks, and communities reveal themselves. Your work community, your family, your activist scene, etc.
If you are in an anarchist community that will probably be abundantly clear from all your minor network connections like going to the same band and knowing the same people as other anarchists.
Even if you never liked an anarchist facebook page or pressed ‘going’ on an anarchist facebook event,
your network is hard to hide.
so
Patterns: whether you realize it or not, you have a data pattern. Maybe you wake up around 10 AM and check your Signal, you use the wifi at your favourite bar on most Sunday nights until about midnight, you use your public transport card to get to class every Monday afternoon. You spend on average 1 hour on Tumblr twice a day. If you have a daily job you will have a very consistent pattern, if you do not your pattern might be more flexible but you have a pattern. If someone wanted to gather your data pattern, they could very easily do so because your pattern is in the metadata.
(Quick side note: in the context of online activity, ‘content’ means ‘the message you send’ and metadata means ‘everything other than the content. So, for example, if you send your friend a text about lunch, the content might be “Let’s go out for lunch” and the metadata is “Message send at 14.32 from phone 0478239055 to phone 079726823 using Signal”).
If you use encryption apps and are not already a suspect, a lot of your online content is not available to most intelligence services, but your metadata is very badly protected by technology and very badly protected by the law. It’s easy to snatch up and analyse to form a pattern.
Now, let’s say you commit a crime, the kind that would result in some serious research. Let’s say that on Sunday night 3 AM, you are your friends decide to go out and burn down a nazi’s house. It’s obvious anarchists are behind it but there are no other clues. You are careful not to communicate about this near technology and you do not leave physical traces.
But because you commited the crime, your metadata will vary strongly from your usual rhythm: you stay at the bar until 2 AM to wait for your friends, you do not wake up at 10 AM in the morning so you do not check your Signal or Tumblr until 1 PM. You do not go to class. Your metadata pattern is very different from your usual pattern and so is that of your friends. If one of you is clumsy, they might generate a super suspicious metadata signal like a phone being switched off at 2.30 AM and activated at 4AM. You wouldn’t be the first.
If I wanted to solve this crime using data analysis, what I would do is:
- let a piece of software run a pattern analysis of the local anarchist scene to generate the 300 people most connected to the anarchist scene.
- let a second piece of software analyse the metadata patterns of those 300 people over the last months and identify the biggest metadata variations around Sunday night
- Illiminate pattern variations with an obvious cause (people who are on vacation, people who are in the hospital, people who lost their job, etc).
- Do indepth research into the ones that remain.
Which is how, out of a massive amount of people that I couldn’t possible all isten to at the same time, I could identify a few to monitor closely. This is how I could find and catch you.
So
How to not get caught: know your network + know your pattern. In the case of the crime above: leave the bar at midnight, return home and put your phone on your bedside table. Check the apps you check before going to bed and set your alarm to 10AM. Return to the bar without your phone. Commit the crime. Wake up at 10AM and check your Signal. Drag yourself to class or ask a comrade to make the trip with your travel card and do not use technology in your home while the comrade is taking your travel card to class. Stick to your pattern. Never ever turn off your phone.
Finally: keep adapting. As technology changes, more information is becoming available, including data you have very little control over. Smart tv’s are one example. Data analysis projects are also currently using license plate reading software on security footage to map the travel patterns of cars. They may soon be ready to do the same with face recognition. More information means more accurate data analysis. Which means you will need to adapt your counter measuresif you want to hide something.
QueerAnarchism 