The loophole to get round minimum audiences is easy - you use the target email address and add eg 99 people of the opposite gender. Then you set the audience ad to the target's gender. Though in principle Facebook should be able to check the effective reach of the ad and stop that.
Imagine a future where one employee is assigned per person to interview and understand exactly what you need. They could just order things before you ask!
Not hard to imagine with advances in machine learning. I'd guess you will almost certainly have at least the equivalent cognitive power of one person monitoring your purchasing needs within the next 20 years.
> The paper’s authors point out a number of cases where nanotargeting was used in a deceptive manner, including the time in 2017 that UK Labour politician Jeremy Corbyn, then leader of the government’s opposition party, decreed that Labour should run a Facebook ad campaign to encourage voter registration.
> Labour party chiefs disapproved the idea, but rather than entering into conflict, simply implemented a £5000 ad campaign designed to only target Corbyn and his associates, as well as a select number of sympathetic journalists. No-one else saw those ads.
I'm not going to lie, that's a pretty impressive way to get someone off of your back.
> The research also suggests that ‘nanotargeting’ users in this way is not only cheap, but occasionally free, since Facebook often will not charge an advertiser for an underserved campaign (i.e. a campaign that only reached one person).
Not free for long, I imagine, now that this is published
the ones I read were people getting their dream jobs because they were smart enough to target an ad about how smart they were and deserved the job to the person doing the hiring.
Probably not. I guarantee one or several really smart people looking at the targeting problem would have considered it during the initial research effort. "I'm going to target X people with this, what are the cost models when X=1, 100, 1K, 1M targets?"
As commenters have noted this has been done before. Likely, the reason why this can't be "fixed" is that it's a result of the curse of dimensionality [1] in the statistical sense: more dimensions make it exponentially easier to separate the space.
This quote from the research suggest it as well, the error rate keep being divided by two while adding dimension (=interests) linearly:
> the formula estimated that ’12, 18, 22, and 27 random interests make a user unique on FB with a probability of 50%, 80%, 90%, and 95%, respectively’.
I would be pretty easy to set a lower limit on the number of people targeted by an ad. If the target audience is smaller than, say, 500 people, don't let that target be chosen.
"In 2018 an AdNews study established that on average, Facebook algorithmically assigns 357 interests per user, out of which 134 were rated as ‘accurate’."
And I imagine variability to be huge in this case. Also average may not 'mean' much as fat tails may cause it to look better/worse that what it actually is.
It's under attack because tyrants masquerading as bureacrats and politicians want to know everything about everyone so they can control everyone. Facebook has a lot of this data and uses it for advertising products and services and adding new features, and sometimes their algorithms can get out of hand the same way cliques on Reddit or Twitter get out of hand when people get brainwashed by doom-scrolling and clickbait.
It's about power and control. These politicians don't actually care about people beyond that, or even about what Facebook is really up to. It's a smear campaign to destroy a legitimate platform that unites people and connects people, because that is contra to what the powers that be want. They don't want that unity. They want fragmentation and isolation in society so everything can decline while they rise in power.
Beware of nanozuks!!! Seriously, what is so malicious about delivering ads based on your personal preferences? Do I need to be targeted by ads that the rest of my demographic prefers?
There are so many ways to abuse micro-targeting. Say hypothetically you’re a private investigator and you have a target that you want to get into their email. You could look at their Facebook page to find their favorite band, food, and travel destination. With that info plus their zip code you could make ads that will just target them personally. Offer free tickets to their favorite band on a landing page with registration, then store their commonly reused password in plain text-done.
This type of ‘nano-targeting’ was the strategy that Cambridge Analytica used to discourage millions of Hillary Clinton supporters from voting. Its been used for countless forms of abuse, manipulation, and data gathering, where even impression counts can be revealing, no clicks needed.
Tragic role of Russian nanozucks in 2016 presidential election cannot be overstated. When did we start believing that handing out millions of flyers is the same thing as selling millions of cars?
It's pretty bad to run campaigns that discourage voting. Accepted practice is to encourage to vote for someone, and not discouraging participation completely.
https://ghostinfluence.com/the-ultimate-retaliation-pranking...
The loophole to get round minimum audiences is easy - you use the target email address and add eg 99 people of the opposite gender. Then you set the audience ad to the target's gender. Though in principle Facebook should be able to check the effective reach of the ad and stop that.