Meta is about to chat with the AI assistant portion of the ad machine, the company announced Wednesday. Starting December 16th, Conversations with Meta AI – The company's chatbots embedded in Facebook, Instagram, Messenger and even the new Ray-Ban display smart glasses will be used to determine the ads and recommendations that will appear in your feed.
The company will officially notify users of the changes on October 7th. There is no opt-out. If you don't want your chatbot conversations to affect your ads, the only option is not to use meta AI at all.
Emily Bender, a linguist at the University of Washington and widely cited co-author. “Probabilistic Parrot is a paper on the risks of large-scale language models (LLMS). luck The company is blurring dangerous lines.
“They are already targeting ads by ploughing your clicks and posts. Now, they're mining conversations with chatbots,” the vendor said. “The next obvious concern is whether the chatbot itself will start tweaking people to disclose more targeted information.”
Vendors claim they are being monitored under the guise of personalization and have the unprecedented ability to extract personal information from users.
“Previously, the Meta system saw who you were involved and what your community was doing. Now it's directly: what are you saying to the company?” she said. “And of course they can combine it with all the other data they already have.”
The vendor says that Meta uses what is called the “illusion of privacy.” People are often in chatbots and are confident about not posting publicly, and have settled into the sense that AI is a neutral listener.
“There's this illusion of privacy. In fact, what you're doing is providing this data to the company,” she said.
However, Meta describes this update as a “natural progression” of its personalization strategy.
Meta's privacy and data policy manager Christy Harris told reporters during a media briefing that people have already assumed that chatbot interactions are being fed into ad targeting.
“We want to be very transparent about this data and raise our heads before we actually start using it in new ways. CNBC.
Harris provided a hilarious example: Ask Meta AI about family vacation plans and you might see more family travel reels in your feed, along with hotel ads. These interactions are treated as new advertising signals whether they are entered into the phone or processed through the microphone on Meta Ray-Ban glasses.
Meta said his current AI work is currently focusing on “building a great consumer experience,” emphasizing that no ads will appear in Chatbot's conversations themselves.
The company also pointed to existing privacy tools. People can “reset or fix AI” via settings, and data retention follows Meta's broader privacy policy.
Meta has poured billions into the generation infrastructure, competed for “Superintelligence” and promised that current Meta AI assistants will help users generate images, draft messages, and plan their day. According to the company, the assistant already has over billions of monthly active users, but that figure includes activity across the entire suite of apps, not just standalone services.
CEO Mark Zuckerberg has made it clear about Endgame. Conversation AI must pay for itself. In May, he said Meta AI will ultimately generate revenue through either advertising or subscription services. Wednesday's announcement is the first major step towards making AI chat a part of Meta's core advertising business.
Meta did not respond immediately luckRequest a comment.
Risk for young users
The implications of this change across meta services can be particularly serious for teenagers and young adults who make up a large part of Instagram's user base and are increasingly attracted to their AI peers.
Vendors call these chatbots “scourges” and warn that framed friendly, always available companions are harmful.
“We've seen people die because of that,” she said, referring to reported cases of people who have been hurt by chatbot interactions. “And let's see how we can make that even more problematic by simply adding an ad to that mix.”
The more she engages AI assistants in the advertising business, the stronger the incentives are and the more incentives they are to maximize engagement rather than helping users.
“Perhaps it will also add to the financial incentives for meta to continue chatting with chatbots. This will optimize engagement, one of the vectors for harm,” she said.
Meta rebutted that protection for young people extends to AI interactions.
“On Instagram teenage accounts, teens are defaulted to the strictest settings of sensitive content control, making them even less likely to recommend sensitive content. Teens under the age of 16 cannot change this setting without parental permission.
