After Meta CEO Connor Hayes announced plans to launch AI-powered accounts on Facebook and Instagram that are indistinguishable from humans, reporters found dozens of similar bots from the company’s early experiments. Real users didn’t want to interact with them because they were little different from accounts that ask to wish AI pictures a happy birthday.
After Meta CEO Connor Hayes announced plans to launch AI-powered accounts on Facebook and Instagram that are indistinguishable from humans, reporters found dozens of similar bots from the company’s early experiments. Real users didn’t want to interact with them because they were little different from accounts that ask to wish AI pictures a happy birthday.
A conspiracy theory that claims that the Internet after the 2010s consists mostly of bot activity and automatically generated content manipulated by algorithms, marginalizing organic human activity.
Proponents of the theory believe that these bots were created intentionally to help manipulate algorithms and boost search results to ultimately manipulate real users.
Recently, Hayes revealed that Meta is going to launch AI character profiles on Instagram and Facebook that «will exist on our platforms just like accounts… they will have a bio and profile picture, and will be able to generate and share AI-generated content on the platform.»
As 404 Media writes, this is another signal that the social network, which supposedly consists of people and is designed for people to communicate with each other, is once again relying on bots designed for the specific purpose of polluting its platforms with trashy content generated by artificial intelligence, like spammers.
But the AI-generated Instagram and Facebook profiles that are now going viral have been on the platform for over a year, and all of them stopped posting 10 months ago after being almost universally ignored by users. Many of the AI-generated profiles announced by Meta have been deleted entirely, and those that remain haven’t posted new content since April 2024, though their chat feature continues to work.
Meta commented that these accounts were created as part of a test that was launched in 2023. They were run by humans and were part of an early experiment the company was conducting with AI characters. «We discovered a bug that was affecting people’s ability to block these AIs, and we are removing these accounts to address the issue,» Meta said.
Apparently, Meta’s AI primarily creates the type of spam that has taken over all of its social networks and turned into a joke. They also show that Meta is not very good at this task and that users do not want it. The complete failure of Meta’s AI profiles confirms once again that people do not come to social networks to interact with bots, but Meta is obsessed with algorithmically pushing such content.
The profiles that remain online are ridiculous caricatures of «people» who post absurd, offensive, and horrific AI content that is in some cases indistinguishable from user-generated spam that pollutes a platform built by enterprising people in India to collect a tiny slice of the advertising revenue that Meta receives.
The most notable account was a bot named Liv, which portrays «a proud black queer mother of two and a truth-teller, your true source of life’s ups and downs.» Her account posted AI-generated photos of children (with traditional hand mistakes) and posts about fake «coat drives» that never happened.
Other Meta bots found included:
Even from these brief descriptions, it is clear why Meta AI characters were not popular and people did not like them. They are only capable of publishing completely banal content with a minimum of usefulness and meaning.
Recall that previously dev.ua investigated how and why Facebook was flooded with inappropriate AI photos of elderly people, newlyweds, Ukrainian military personnel, and other people asking them to «bless», «wish them a happy birthday», or «send some love».
