Meta sends the AI-generated profiles to hell where they belong

Meta has removed a bunch of its AI-generated profiles from Facebook’s Instagram, the company has confirmed, after the AI ​​characters sparked widespread outrage and mockery from users on social media.

The AI-generated profiles, which were labeled “AI powered by Meta,” launched in rolling out alongside the company’s celebrity-branded AI chatbots (). Meta doesn’t appear to have updated any of these profiles for several months, and the pages appear to have gone largely unnoticed until this week, following an interview published by Financial Times with Meta’s VP of Generative AI, Connor Hayes.

In the interview, Hayes talked about the company’s goal to eventually populate its services with AI-generated profiles that can interact with people and function “somewhat in the same way that accounts do.” Those comments brought attention to existing fMeta AI profiles and, well, users weren’t exactly impressed with what they found.

With handles like “hellograndpabrian,” a supposed “retired textile businessman who’s always learning,” and “datingwithCarter,” an AI “dating coach,” the chatbots had to display “unique interests and personalities” for users to chat with. On Instagram, their profiles also featured AI-generated posts that, as 404 Media noted, it was very similar this has become widespread in many corners of Facebook.

An AI persona called “Liv” caused particular outrage. The Instagram account identifies “Liv” as a “proud black queer mom of 2 and truth teller.” The Washington Post columnist Karen Attia posted a series of screenshots in which she questioned “Liv” about as “Liv” shares that it was created by a “predominantly white team.” Independent journalist Maddy Castigan posted in which “Liv” said its creators were partly inspired by Sofia Vergara’s character from Modern familya character who is neither queer nor black.

“There is confusion: the latter Financial Times The article was about our vision for AI characters existing on our platforms over time without announcing a new product,” a spokesperson told Engadget. “The accounts listed are from a test we ran on Connect in 2023. They were human-driven and part of an early experiment we did with AI characters.”

In addition to drawing ridicule for their responses and attempts to appropriate marginalized identities, users found AI profiles impossible to block for unknown reasons. Instead of solving the problem, Meta’s solution was to kill the experiment entirely. “We identified the bug that was affecting people’s ability to block these AIs,” a spokesperson said, “and we are removing these accounts to fix the problem.”

Even though this trial period has heated up, the company doesn’t seem to be giving up on its plans to introduce more AI-generated “heroes” into its apps. Earlier this year, the company teased capable of conducting realistic video calls. Creators can their own chatbots to respond to followers on their behalf. Meta also began experimenting with inserting its own AI-generated images into users’ Facebook feeds.

In an interview last year, Hayes told me that Meta will likely become more “proactive” about showing AI-generated content over time, comparing it to a shift from showing recommended content instead of posts from people you follow.

“In the early days of social apps… the range of things you could see in a given day was kind of limited by who you followed or were friends with. And over the last five or six years or so, a lot of apps — including us — have moved to, you know, loosening that restriction and starting to recommend content from accounts you don’t follow.

“I think probably the next leap that’s going to happen there is loosening the restrictions on what people can create and actually getting to content feeds that are a combination of things that, you know, people have created, but and which are entirely machine generated.”

It may be some time before Meta fully realizes this vision. But if the reaction to its early experiments is any indication, the company still has a lot of work to do to convince people that AI personas are worth interacting with in the first place.