Surveillance capitalism doesn't care at all about the accuracy of the data that they have. There is literally no such thing as "garbage data" to data brokers. It's all just data, and all of it is valuable.
That's why projects that claim to "pollute" your browsing history like RuinMyHistory, noiszy, adnauseam, and TrackMeNot are not only pointless but also dangerous.
The data being collected about you will always be used against you, no matter if it is accurate or not. If your browser randomly browses to webpages that gets "this person is a muslim" or "this person is gay" added to your dossier it doesn't matter if it's true or not, when your next would-be employer or would-be landlord who hates muslims or gay people uses a data broker for "background checking" and sees that, you're not getting the job/apartment. They won't tell you why, you'll just be rejected/ghosted.
If you're a 40 year old man, but your browser add-on convinces a data broker that you're a 34 year old woman seeking an abortion, that data can still cause you end up the target of a lawsuit in Texas and it will take a non-zero amount of time and money to clear that up.
If someone in your zip code kills someone using a certain type of plant, or household cleaner/chemical, or medication and your add-on has been browsing sites about that thing, you can end up on the police's suspect list.
If you only make $30,000 a year but your add-on searches for yachts and expensive jewelry often enough to convince a data broker that you've got tons of money then it doesn't matter that the data is wrong, the next time you try to book a hotel or order something online you can still be charged a lot more than you would have been charged otherwise.
Handing extra fake data to people whose only goal is to use data against you is just handing them more ammunition. It doesn't matter if it's "garbage" to you, it's still something they can and will eventually use against you. You cannot know what will prejudice someone against you. The more data is in your dossier, the more opportunity there is that you'll meet the right (or wrong) criteria.
No data broker is going to look over your dossier and see that there's inconsistencies and go "Damn it! This genius has ruined my data! Now I have to throw all this data away as it is now worthless!" They aren't even going to look over your dossier. They're going to get paid to hand over a list of people flagged as being 'X' and your name/address/identity will show up along with everyone else flagged as being 'X' even if your name gets pulled up again when someone else pays that same data broker for 'Y' which is the opposite of 'X'. The data broker gets paid either way.
If the accuracy of the data are not intrinsic to their value, then data brokers could literally just manufacture data. “Is this person X?” flips coin “yes!” Who are you to say otherwise? We have a vast network of blah blah data sources and advanced AI inferencing. They would absolutely do this if they could get away with it because it is dramatically cheaper.
Targeted ads exist. People find them “creepy,” which implies that they are targeted based on factual data. Therefore, we know that data brokers are taking the more expensive route and collecting factual data (or striving to). They would not do this without a profit motive. Perhaps their data are being compared with a competitor’s to enforce quality… we don’t really know. But we know that they value the quality of their data because their customers do. Consequently, it must be the case that deliberately polluting their data devalues their product and erodes their business model over time.
> If the accuracy of the data are not intrinsic to their value, then data brokers could literally just manufacture data. “Is this person X?” flips coin “yes!” Who are you to say otherwise? We have a vast network of blah blah data sources and advanced AI inferencing. They would absolutely do this if they could get away with it because it is dramatically cheaper.
Measuring the effectiveness of advertising has always been difficult. Targeted ads can be very effective at times, and at others do no better (or worse) than chance. The many clear failures of targeted advertising hasn't hurt the industry though and it isn't likely to either. Now companies are advertising AI as the new thing to increase their accuracy. How well that works for them remains to be seen.
All of these -- barring illegal discrimination -- are actual problems that will come to light when it turns out the information is incorrect.
> If you're a 40 year old man, but your browser add-on convinces a data broker that you're a 34 year old woman seeking an abortion, that data can still cause you end up the target of a lawsuit in Texas and it will take a non-zero amount of time and money to clear that up.
It will make the data broker look bad when the prosecutor finds out that they fabricated my abortion. It might be annoying and stressful for me but I'm sure I'll get through it. It seems like I would have the same response to everything except the illegal discrimination as mentioned. If this is the result of me running the extension, I only see upsides.
> All of these -- barring illegal discrimination -- are actual problems that will come to light when it turns out the information is incorrect.
If I need a hotel and the place I'm booking decides to change me more than they would have otherwise because they're mistaken about my finances, I'm still getting charged more. None of that comes to light.
When a store tells me that their return policy is "next day only, with receipt" but they tell the next person in line it's "30 days no questions asked" all because their "consumer reputation service" told them I was unreliable when I'm not, I'm still stuck with their shitty return policy for "bad" customers. None of that will ever come to light.
When my health insurance company jacks up my premiums because a data broker told them that I've been spending more time at fast food restaurants, I'm never told that's what happened, I just get a bigger bill. Nothing ever comes to light.
When the police arrest me and question me because of my search history, maybe the truth comes to light, but not without significant costs to me.
Most of the time when people use the data that's been collected about you as a result of surveillance capitalism you have no idea that it even happened or why. You're just charged more money than you would have been, or you aren't offered opportunities you would have been given, or you're just rejected for something you wanted, etc. Nobody tells you why. There's not an investigation into how it happened. There is no transparency and there is zero accountability for errors.
> It will make the data broker look bad when the prosecutor finds out that they fabricated my abortion.
When have you ever heard of a data broker taking a huge hit to their reputation because they have inaccurate data? It doesn't happen. What data broker has a great reputation in the first place? Everyone using data brokers knows that the data is not 100% reliable. It doesn't matter. It's usually just a numbers game. Even when it's only for an advertisement, they know that not everyone they're targeting is going to buy something. That doesn't matter to them as long as some percentage does.
Do we know whether any of this stuff is actually happening, in reality, to actual people, based on some IP address's history of clicking ads? Any concrete examples you can link to?
Data brokers get their information from all kinds of sources. There is no complete breakdown on where it all comes from in every instance that the data is used, part of the problem with surveillance capitalism is that there is zero transparency and near zero accountability, but yes, data brokers do collect your browsing history and that includes what ads you view/click
As for examples of that data being used "in reality, to actual people" you might find some good info in these links:
Employers and landlords using data brokers for hiring/rental decisions:
Keep in mind that this is a rapidly growing space.
Travel sites, retailers, even grocery stores have been looking into how to use this kind of data to set the prices of their goods on an individual basis to make sure that they can squeeze as much money out of you as possible. The main thing holding them back so far is that consumers view discriminatory pricing as unfair, but they've been working hard for a long time to change that view. If you happen to find a place that requires you to scan a QR code to see prices or get a menu, you might want to check with the people around you to make sure everyone is paying the same price.
> when your next would-be employer or would-be landlord ... uses a data broker for "background checking"
Knowing an SSN (US Social Security Number) was used like a password by banks and all kinds of organizations, but not anymore. Things change, albeit slowly, when society acknowledges mistakes. All of your examples rely on the data buyer trusting its accuracy. If enough people pollute the data, then it'll have no value. The data brokers won't be able to sell it because society will know that its garbage.
The vast majority of the time, the people buying the data know it isn't 100% accurate and they also don't care. They're usually not looking at individuals, they're looking at large groups of people who (probably) meet whatever criteria they've set. If they get it wrong a bunch of times who cares as long as the other times it works. Better than chance is good enough for them.
Data brokers can always count on there being some people who are misled into thinking that their data is far more accurate than it is. They're trying to convince everyone right now that AI is the golden solution that makes them super trustworthy compared to their previous failings. Tomorrow it'll probably be "quantum something" or "super surveillance" or some other gimmick. They really don't have to care. They'll always be able to sell their stuff.
Data brokers will always have the police and the government buying up their data too because the government is happy to just suck up everything they can and will figure it all out later. They aren't concerned with accuracy either. That's how people get arrested for just riding their bike past houses that got robbed. They can make all the mistakes they want and it doesn't hurt them any, even when it's a huge problem for the people caught up by lazy policing.
I promise you that no browser add-on is going to collapse the targeted ad industry or bring an end to surveillance capitalism. Giving people more data to use against you is a bad idea.
> Surveillance capitalism doesn't care at all about the accuracy of the data that they have
It’s clear and obvious that they do. If the data was made up, they wouldn’t be able to serve effective ads.
> If you're a 40 year old man, but your browser add-on convinces a data broker that you're a 34 year old woman seeking an abortion, that data can still cause you end up the target of a lawsuit in Texas and it will take a non-zero amount of time and money to clear that up
This situation would absolutely never happen, and I think it’s blatant fear mongering.
> If the data was made up, they wouldn’t be able to serve effective ads.
The data being collected about you isn't about advertising. It's used for an ever increasing number of things that impact your real life including things like how much you get charged when you buy things, what services you're told exist or are eligible for, how long you get left on hold, what policies a company will tell you they have, and who will hire you. The data being collected about you can be used against you by police, or in courtrooms, and in custody/divorce hearings.
Even when the data is used for advertising (along with scam attempts, the manipulation of your opinion, and political propaganda) "effectiveness" is a very uncertain thing. No one expects that everyone they target with a campaign will bite. The effectiveness and accuracy of targeted ads isn't exactly certain to begin with.
> With just one parameter - gender - the data is only 42% accurate. That is less accurate than if you just did “spray and pray” with no targeting at all — i.e. you would have still hit the right gender 50% of the time. With two parameters - gender plus age - the accuracy is down to an average of 24%. Some data brokers were far worse, with single digit percent accuracy. Third party profiling of audiences is so inaccurate, it’s better to save your money and do “spray and pray” instead. (https://www.forbes.com/sites/augustinefou/2021/04/19/ad-rele...)
> This situation would absolutely never happen, and I think it’s blatant fear mongering.
On the contrary, Texas in particular has gone out if its way to incentivize its citizens reporting other citizens for getting abortions. The law in that state creates an incentive for someone with the ability to do this, to do so.
What's special about the situation in Texas right now where humans at large will not follow the incentives placed in front of them for the first time in human history?
> Texas in particular has gone out if its way to incentivize its citizens reporting other citizens for getting abortions
That’s true, but if you, a biological man, were sued for having an abortion, this would be immediately thrown out. Indeed, even civil suits with their lower standard of evidence require more than a simple search history from data brokers.
Of course the case would be thrown out. But would it be thrown out without me having to do anything, or would it cause me to spend time and possibly money dealing with it? Especially if I had an androgynous name, there's a very real chance that I would have to physically show up in a courtroom somewhere to point out that I do not have a uterus.
