Researchers show Facebook’s ad tools can target a single user
techcrunch.com
techcrunch.com
Labour staffers working to prevent their own party winning under him were not posting the adverts they were meant to, but were able to use geofencing and other specifiers so it looked like everything was it should be when he was browsing.
https://www.thetimes.co.uk/article/labour-hq-used-facebook-a...
https://ghostinfluence.com/the-ultimate-retaliation-pranking...
Also, can't read Techcrunch articles because they're blocked (by NextDNS) due to redirecting through https://guce.advertising.com/
Yes, it was discovered 7 years ago (by many) and Facebook has continually lied about it being patched.
Here's a more recent overview of all their lies regarding micro-targeting:
https://thetyee.ca/News/2019/05/07/Facebook-Privacy-Busting-...
We’ve known unambiguously about the impact big business had had on our climate, yet it took decades for governments to act decisively - mine still hasn’t.
We knew about the hazards of smoking for decades before government acted decisively, despite the enormous costs to society. In many cases these companies continue to be allowed to sell their shit, despite knowing that it’s not just addictive, but also toxic.
Asbestos, leaded fuel, road safety, and many other large problems were left unregulated for decades, and in all cases a small number of corporations profited - often marginally - at the expense of thousands or even millions of individuals.
And now we see the same with Facebook. We’ve known for more than a decade that Facebook is a toxic waste pit of information.
There is much that could be done, and much more quickly, but our politicians sit on their fat arses and do Jack shit.
Without effective democracy we have no way to resolve these problems because the whole reason we have a democracy in the first place is to ensure these powerful bastards work in the best interest of everyone.
If King Zuckerberg is more powerful than his parliament of lapdogs, then democracy is dead.
/rant
While you can target a SINGLE person, the targeting doesn't tell you who that person is. So, I can target my wife - but I already knew who she was. Also, I can target someone using 18 parameters, but I don't actually know who that person is unless they click on my add and tell me "this is who I am".
So: does this violate privacy laws?
The concern that immediately comes to mind is that this nanotargeting ability could be abused by (just to name a few) spiteful ex-partners, stalkers, internet trolls, and generally vengeful unpleasant people.
Interestingly, there is a different way to go about it.
Assume you create an ad for diamonds, and target it to people who are into adultery. Your wife, who likes diamonds, sees the ad, clicks it, and fills in her information on the diamonds website (a mock-up you built). Now again, you know that your wife is interested in adultery. But she is unaware that she even revealed this information in her transaction with the "diamond" website. She wrongfully assumed that all the private information she shared with FB stayed on FB's servers only. Note that this leak also exists in Google.
Unless there is proof they DON'T do this, like some law that is clearly enforced, or some piece of code that mathematically prevents it from doing so... I think it is more than safe to assume they can and they have violated privacy concerns.
Obviously a gross oversimplification, but to a first approximation ads ranking, pacing, and auction drive ads revenue. Ads targeting drives careers, at least at FB at least as of a few years ago.
The creepy factor does way more damage to FB’s business than the bells and whistles that the buyers of inventory use adds to it.
When something happens on Google, people feel like it happened in their living room. When something happens on FB, people feel like it happened in their bedroom.
They helped to create a genocide in Burma. If people didn't leave after that then I doubt any "creep factor" is going to hurt their business.
However, regardless of privacy laws and other issues with advertising, this particular kind of abuse could be prevented by displaying how many people fall into that ad's targeting criteria and how much this advertiser has spent in the last year. Legitimate ads will typically target a large amount of people and would've spent significant amounts, where as "gaslighting" ads like these would typically reach very few people and the advertiser would've spent a relatively low amount in comparison.
Facebook incited a genocide in Myanmar. PR is not in their list of worries.
The random selection could be uniform or it could be weighted by the global popularity of the interests. Presumably it's uniform. But if so, you could challenge the reasonableness of that assumption since it would be more likely to know a more common interest of a user that one in the tail of the distribution.
For example, suppose I have three interests, two of which are very common and one of which is extremely niche. If you select two of these interests uniformly at random, chances are you'll get the niche one and be highly likely to identify me. OTOH if you randomly select weighted by the global popularity of the three interests, chances are that you'll get the two common interests and be very unlikely to identify me.
Why would we get addicted to such evil? What's wrong with our minds, are we doing something wrong and can we teach next generations to be better than ourselves?