Facebook's owners have patented an AI tool that could continue posting on behalf of a user who has died. There's no sign it plans on using the technology but it has still raised some significant ethical questions.

Meta, which owns Facebook, Instagram and WhatsApp, applied for the patent in 2023. It was finally granted in December and then spotted by Business Insider. (Source: businessinsider.com)

The patent covers a technology that would allow the company to simulate a user's activity when they are unable to post for an extended period. That could be because of illness, time away for work, or if the user dies.

Auto-Likes and Comments

Bizarrely, the patent application says the tech would solve the problem of a user's absence affecting other users who followed them on a social media site. It makes the somewhat understated point that "The impact on the [followers] is much more severe and permanent if that user is deceased and can never return to the social networking platform."

Although AI-generated posts would seem the most obvious use case, the patent suggests it would be more about simulating other activity such as leaving comments on other people's posts and "liking" content. The idea would be to analyze past behavior and then act how the algorithm determined the user would have responded to other content.

"No Plans" to Develop Concept

What value that brings either the user (if they are still alive), or the people whose content are receiving automated "likes" and comments is not entirely clear.

Meta says that it has "no plans to move forward" with the technology and that filing a patent does not mean it will turn a concept into a working technology. It's possible it began working on the idea and has since concluded it would be poorly received. It's also possible its more of a defensive patent that means it could block other companies from using the technology. (Source: dexerto.com)

Facebook Acquires AI Agent Social Network

The concept of AI bots interacting on social networks is not purely theoretical. Just a few days ago, Meta acquired a startup called Moltbook, which built a social network specifically designed for AI agents. On that platform, automated bots can generate posts, reply to comments and interact with one another much like human users.

According to reports, the Moltbook platform effectively functions as a social network where AI agents hold discussions and produce content autonomously. Meta reportedly brought the startup's founders into its Superintelligence Labs, suggesting the company is exploring how autonomous AI agents might interact on social platforms in the future.

Elements of this idea already exist on other platforms. For example, on X (formerly Twitter), users can mention the platform's AI assistant by typing @grok in a post and asking it to analyze or comment on a tweet. The AI then replies publicly in the conversation thread, demonstrating how automated agents can already participate directly in social media discussions alongside human users.

What's Your Opinion?

Are you surprised by this filing? Do you think Meta will ever use the technology? Is the concept acceptable as long as the user gives (or gave) their express consent?