AI Chatbots may be coming to Facebook soon

Meta could be on the verge of jumping back into the AI chatbot battles—and bringing a former president along with it. A report in the Financial Times says the social media company plans to launch a series of chatbots, each with different personalities, including one that impersonates Abraham Lincoln. Another would offer advice on travel options, using the dialogue of a surfer.

The chatbots could begin to roll out as soon as September, the report claims. (Meta did not reply to a request for comment by Fast Company.)

Besides being a fun novelty for users to play around with—and a reason to spend more time on Meta’s apps—adding the AI chatbots could potentially boost the search functions of the company’s holdings, which include Facebook, Instagram, and Threads. It could also offer recommendations.

Of course, there’s also the potential to use the chatbots to collect even more information about users, their interests, and habits, letting the company better target them with advertising. Snap CEO Evan Spiegel, in June, told the Wall Street Journal that 150 million people have used Snapchat’s chatbot, sending more than 10 billion messages—something he felt could improve ad targeting on the platform.

Meta, of course, has dipped its toe into the AI chatbot waters before—and it didn’t go so well. Roughly a year ago, Meta launched BlenderBot 3, a prototype the company said could discuss just about every subject. Almost immediately, reports emerged that the company bashed its maker (telling one user that Mark Zuckerberg was “a good businessman, but his business practices are not always ethical” and another, “Since deleting Facebook, my life has been much better”).

More concerning, the AI also reportedly made anti-Semitic remarks and declared that the 2020 U.S. presidential election was stolen. Facebook made some rapid changes, but users had already moved on and BlenderBot never found an audience after its disastrous debut.

Of course, there’s also the potential to use the chatbots to collect even more information about users, their interests, and habits, letting the company better target them with advertising. Snap CEO Evan Spiegel, in June, told the Wall Street Journal that 150 million people have used Snapchat’s chatbot, sending more than 10 billion messages—something he felt could improve ad targeting on the platform.

Meta, of course, has dipped its toe into the AI chatbot waters before—and it didn’t go so well. Roughly a year ago, Meta launched BlenderBot 3, a prototype the company said could discuss just about every subject. Almost immediately, reports emerged that the company bashed its maker (telling one user that Mark Zuckerberg was “a good businessman, but his business practices are not always ethical” and another, “Since deleting Facebook, my life has been much better”).

More concerning, the AI also reportedly made anti-Semitic remarks and declared that the 2020 U.S. presidential election was stolen. Facebook made some rapid changes, but users had already moved on and BlenderBot never found an audience after its disastrous debut.

Meta has mostly been focused on the metaverse, but when it laid off workers earlier this year, it also signaled it would reprioritize AI. And in the company’s second-quarter earnings report on July 26, Zuckerberg made it clear that AI would be a big part of Meta’s future.

“I’m going to share more details later this year, but you can imagine lots of ways AI could help people connect and express themselves in our apps: creative tools that make it easier and more fun to share content, agents that act as assistants, coaches, or that can help you interact with businesses and creators, and more,” Zuckerberg said on a call with analysts. “These new products will improve everything that we do across both mobile apps and the metaverse—helping people create worlds and the avatars and objects that inhabit them as well.”

Meta might be the first company to roll out an AI with the simulated personality of a long-dead president, but it’s hardly the first to introduce a chatbot that apes the mannerisms of a real person. Character.AI is an app that lets people “chat” with characters, both fictional and real, including a simulated (and rather bombastic) Elon Musk.