Originally published at: Facebook mistakes coding teacher for animal trader, bans him for life with no appeal | Boing Boing
…
2 Likes
Fuck Facebook, for sure, and all this “the algorithms will do all the work, we’ll make billions!” bullshit. It does make me wonder if there are any underground exotic animal salespeople trying to use the names of coding languages to actively trick ad algorithms into letting them through under the guise of “code tutorials…”
11 Likes
As I’ve said before, the same thing happened to me. As far as I know, I didn’t get mistaken for an animal trader, although I’m sure I’ve used the word Python at some point. It doesn’t impact me in the slightest, because I don’t run any ads on Facebook, but it was very instructive to see how completely without recourse such an algorithmic decision is. At no point was a human involved. Had I actually depended on that account for my livelihood, it would have been devastating.
15 Likes
Again, and again, and again we see the flaws in algorithms and relying on them for ANYTHING that requires ANY values judgements. They are flawed, often built upon racist or sexist data, and are treated as sacrosanct when anyone asks for an explanation. “The algorithm makes no mistakes.”
And this is the technology underlying chaptGPT and other False AI (Fai) systems. And we continue to fail to do anything to regulate these tech douche bags because profits uber alles is the most important value among the elite who run the country and our system of economics.
15 Likes
13 Likes
Speaking of letting AI make judgements on moderation, I keep getting temporarily banned from commenting on instagram and threads for totally innocent things and I can only think it’s because I used certain words. Meanwhile I see hate and disinformation shared on there constantly. It reminds me of the Patton Oswald bit where he points out it’s worse to society to say heartfelt and inclusive things but use the wrong words than it is to be a polite well spoken bigot who uses intellectual language to justify their bigotry. (In my case I didn’t use any bad language but I did make a historical comment about the fact that Walt Disney produced anti-n@zi propaganda in ww2… which I guess was just way too many bad words in a row)
7 Likes
I’m reminded of a premise from Laverne & Shirley in which the computer was, ostensibly, infallible.
1 Like
The rules are simple:
By the standards of AI, you need to be able to simulate some aspect of authenticity,
1 Like
There have been quite a few instances of folks being instantly banned from services like Paypal for unknown reasons that an algorithm suddenly decided that a normal innocent person is up to no good, and there’s apparently no way to undo the ban either. These automated systems seem useful if the ultimate decision is left up to a reviewer but that’s hardly ever the case 
it’s all so stupid.
5 Likes
I think you mean False AI Language systems.
2 Likes
At this point there’s still no real AI, so they’re all False AI, language system or otherwise.
3 Likes
Came here to say the same. Well, the first two words at least.
1 Like
“I recommend Pangolin. It scales up easily.”
3 Likes
I wanted to launch a new acronym.
1 Like
i can merge these two threads, he thought to himself.
4 Likes
This is not a new problem, it’s been a know issue for decades
4 Likes
I can almost understand… on some level I kinda respect rules to create civility. I’d adjust or delete my comment if I had the chance. The weird thing for me is not even knowing until afterwards that I’ve broken a rule and been punished. Like for crying out loud give me a warning and tell me what triggered it. And the complete lack of any explanation or way to appeal. That’s what’s driving me nuts.
1 Like
