Instagram head Adam Mosseri posted on his account on Wednesday to dispel the myth that the social networking giant is actively “listening” to its users surreptitiously, in order to target them with relevant ads. The idea that Meta would secretly turn on the microphones on users’ phones to record their conversations is an age-old conspiracy theory — and one that the company has disputed before.
But, ironically, Mosseri’s new myth-busting claim comes just as Meta has announced it will soon target ads to users across its social apps using data collected from their interactions with its AI products.
In other words, if Meta didn’t need to record your conversations via your microphone before to produce eerily accurate recommendations, it definitely won’t need to now.
On Instagram, Mosseri says he’s had a number of conversations about Meta listening to its users, many of whom can’t believe how well the company’s ad targeting actually works. (Even his wife has brought up the topic, he says.)
By now, most of us have either had the experience ourselves or at least know someone who claims that Meta must have been secretly recording them to know what they were likely to click on. Sometimes, you are only thinking about a topic or product, and then see the content appear in your feed, making it seem as if Meta is a mind reader.
The company has repeatedly disputed these claims, trying to explain that it doesn’t have to record your conversations to make its recommendations so successful. (Mosseri also says that would be a “gross violation of privacy,” but Meta is not a company that typically drives decisions with user privacy in mind.)
Still, the company doesn’t necessarily have to “listen” to users to listen to them.
Techcrunch event
San Francisco
|
October 27-29, 2025
In 2016, Meta (then known as Facebook) published a blog post that outright stated that it didn’t use your phone’s microphone to determine what ads to show users or what content appears in their News Feed. Years later, Meta CEO Mark Zuckerberg testified before Congress, denying once again that the company was collecting users’ audio data for this purpose.
Happy to have something it can deny on the privacy front, just as it’s about to scoop up more data than ever before, Mosseri reiterates these points in his post on Instagram.
He says that, for starters, users would know if their phone’s microphone was on because they would see a light at the top of their screen, and the phone’s battery would drain faster.
Instead, Mosseri explains that the tech giant’s recommendation system is so powerful because of how it works with its advertisers, who share information with the company about who has visited their websites. That information helps Meta target users with relevant ads. In addition, the company shows people ads that it thinks they may be interested in based on what similar people with similar interests are also interested in. This algorithm-based ad tech has made Meta a money-printing machine over the years.
Now, Meta is going to leverage AI to make these ad targeting decisions. So if people thought they were being listened to before, it will only get worse. The company said its new privacy policy, which is being released on December 16, will allow it to use data from consumers’ interactions with its AI products in most markets as another signal. And it’s a potentially even more powerful one than the “people who like this also like that” system, given that users are engaging in far more personal conversations with AI chatbots like Meta AI about their interests, ideas, and activities.
Mosseri also points out that sometimes it’s not technology alone that’s driving the hyper-accurate recommendations — it’s either just a coincidence or a bit of human psychology at play.
“You might have actually seen that ad before you had the conversation and not realized it,” he points out. “We scroll quickly. We scroll by ads quickly. And sometimes you internalize some of that, and that actually affects what you talk about later,” Mosseri says.
