That-one-thing

In Sweden we have iq.se which is a site informing about drinking responsibly and can help with information. But now they released a bot that is supposed to help students drink more responsibly by sending messages on messenger before, during, and after the party with advice[0].

So I started wondering, won’t the algorithm find that there is a large group connected by this bot where most people drink a lot ( I assume that’s the target demographic ). Isn’t there a chance that it might try to take advantage of their condition. Because we saw that Facebook had created ads to target racists[1], then I think it should be able to figure out how to target people who have drinking or impulse problems. We had the thing about being able to tell when people feel insecure or depressed to target them if they want to (claimed to not do that)[2].

So isn’t there a large chance that something like this bot or AA groups on Facebook might be used to target people when they feel left out, want a drink, feel insecure. Not that the ads will necessarily be directly targeted towards them, just that it might be a big factor?

[0] http://iq.se/bottlier

[1] https://www.nytimes.com/2017/09/20/business/media/facebook-racist-ads.html

[2] https://www.usatoday.com/story/tech/news/2017/05/01/facebook-can-tell-when-teens-feel-insecure-advertiser-target/101158752/