Facebook ditches its creepy, controversial robot – yes, its facial-recognition AI Having last week sidelined the tarnished brand Facebook to conduct business under the name Meta, the social ad biz intends to deactivate at least some of its facial-recognition systems in a few weeks. In a blog post on Tuesday, Jerome Pesenti, VP of artificial intelligence at Facebook, said the social ad platform – still …
As one of the cadre of people who never agreed to use Facebook, I still want them drawn and quartered for enabling facial tagging in photos w/o the consent of the people being tagged. Yes way back at the beginning that's how it worked. They also would spam your email address if one of your friends made the mistake of ever logging into the app on a device with your contact info on it. And of course the only way to opt out was to permanently create a Facebook account and then log in constantly to discover what privacy settings they changed and what new hell they had opted you into. And even more spam.
How much you want to bet the plan to "shutdown" Facial Recognition won't involve disclosing who they tagged(including non-facebook users), what third parties accessed or used the information, or let those people (users or not) review that data, or delete it after the inevitable storm of lawsuits that are coming.
"How much you want to bet the plan to "shutdown" Facial Recognition won't involve disclosing who they tagged(including non-facebook users), what third parties accessed or used the information, or let those people (users or not) review that data, or delete it after the inevitable storm of lawsuits that are coming."
Yeah, I thought that too when I read:
"no longer supply names for people recognized in photos."
I'd be prepared to bet it's all still going on in the background, just not publicly showing the results.
... Facebook has decided that the AI algorithm has had enough training data to ID six or seven 9s of all photos and no longer needs the "help" of the luserbase putting names to faces.
I'll give you even odds that it "accidentally", if quietly, gets turned on again in six to eighteen months to update that data, and stays on until somebody conclusively (and publicly) proves that it is, in fact, on.
As for Facebook deleting the data... bullshit. Facebook has never deleted anything, They aren't going to start now. Anybody who thinks otherwise is a fool. Even if they very publicly go out of their way to "prove" that the data is off their systems ... well, that's what backups are for. And you know they have backups. They may even "prove" they delete the backups. That's what off-site backups are for. And you know they have those, too.
Or maybe they just found it doesn't work well enough and thereby it's useless to sell ads. If it can't identify people well enough it could led to advertise the wrong - and probably offensive in that context - product, and that's really bad for advertisers.
They are deleting the templates - still they have all the raw data to recreate them as they wish.
wrong - and probably offensive in that context - product
All unwanted advertising is offensive. It steals my time, my bandwidth, my power, and my power to try and convince me to purchase your product/service.
I'm not opposed to advertising in a paid-for public arena. I am opposed to targetted advertising and by implication the data collection and collation that is required to make it work (if it does indeed work).
It depends on what you can make advertiser believe or not. If face recognition is slapped here and there for huge mistakes especially in some categories, advertiser will know it - and maybe can also see it themselves. While how well targeting works it's more difficult to know - especially when people don't buy directly online on the spot.
"Big companies presumably listen to their consumers and respond to privacy laws"
Rubbish. Big companies pay to get laws bent to their will (or just ignored). And as been said ad nausium here, Facebook's consumers/customers are the advertising companies, not the peons who post their life story on it.
Internet history is littered with examples "here's something cool... oh no it's being taken advantage of". This is another. As a regular user it is so cool to be able to search for photos of X, or have photos auto-tagged to save me doing it (who can be bothered).
But it so clearly has scope for misuse. And then the PC brigade wade in with claims it's racist and that's the nail in the coffin.
It seems everything that in SciFi is exciting, is only ever bad in real life.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
