>>1501405>>1501390They definitely detect AI generated images. Very easy to detect, literally 5 lines of javascript. Countless of methods and libraries for this purpose. They scan all images for everything - content, exif, people in them and their relations to each other etc... Facebook has extensive backend system for keeping track of what gets uploaded when, who is on it, sentiment and emotions of the people present etc etc.
>>1501351I'm not certain, however, if it's anything like Reddit or other similar anti-spam/anti-bot measures on big companies, it's something along the lines of:
>no phone numbercan you even register without a phone number? even if you can, doing so is definitely a trigger
>VPNthe system keeps tabs on VPN addresses in the world. it's an instant trigger. nothing escapes it. they have contracts with ISPs and VPN providers to share that info with them
>temp mailinstant trigger. no temp mail provider escapes that
>brand new email addressthe system knows if the email address is new and most big email providers report that information. instant trigger
>behaviorthere is definitely an intelligent system in place keeping track of new user behaviour. if your behavior doesn't fit the parameters it will trigger it. reddit has this: if you make a new account and start browsing some obscure subreddits, searching for exact things in the searchbox, posting too much or following accounts or chatting in DMs, it shadowbans the account.
if there is a way to make an account without face scan, it should involve not triggering their mechanisms. don't use vpn, make a new email and use it for a week or two, register to some porn sites or whatever, let it rest and cultivate. if it asks you, use a phone number. your own or buy a prepaid or something. when you register don't go directly to marketplace and search for shit. act like a new user would act. scroll the timeline for a minute, close it, return the same night for a few minutes etc etc...
