According to Gizmodo, Meta updated its privacy policy last month to explicitly state that data from user interactions with its AI services will now be used for targeted ads across Facebook, Instagram, and WhatsApp. The policy now covers “Prompts that can include questions, messages, media and other information” shared with “AI at Meta,” which includes the Meta AI chatbot and tools for Ray-Ban smart glasses. The company first alerted users to this change back in October via a press release, giving the example that a chat about hiking could lead to ads for hiking boots. This shift has already drawn a formal backlash, with a coalition of 36 privacy and consumer protection groups sending a letter to the Federal Trade Commission (FTC) at the end of October urging an investigation. They argue the move is an aggressive expansion of surveillance-based marketing. Meta says over 1 billion people use Meta AI every month.
The User Impact: It’s Creepy Personal
Here’s the thing: this isn’t some theoretical new data stream. It’s your casual, possibly private, conversations with an AI. You ask Meta AI for a recipe, vent about a bad day, or brainstorm gift ideas for your partner. Now, that raw, contextual data is fair game for the ad machine. Meta frames it as a way to show you “more of the content you’re actually interested in,” but the line between helpful and invasive is incredibly thin. And they’re doing it without a basic opt-out. So your most exploratory, off-the-cuff chats—the ones where you might reveal nascent interests or vulnerabilities—could become the blueprint for what you see in your feed. It makes the already-opaque “why am I seeing this ad?” algorithm even more inscrutable.
The Bigger Picture: A Deliberate Normalization
The advocacy groups nailed it in their letter to the FTC: this is about normalization. Meta is strategically expanding what we accept as usable data for ads. First it was your likes, then your location, then your off-app browsing. Now it’s the content of your direct chats with an AI assistant. Each step makes the next one slightly less shocking. They’re betting that the convenience of a slightly more relevant ad will outweigh the creeping feeling of being constantly listened to. And with a billion monthly users, that’s a massive behavioral dataset they’re now monetizing. The real question is, where does it stop? If AI chat data is in-bounds today, what about data from future, more intimate AI health coaches or financial advisors?
The FTC: Will Anyone Push Back?
So, will the FTC actually do anything? That’s the big unknown. The groups are leaning hard on the agency’s 2019 consent decree with Meta and Section 5 powers against unfair/deceptive practices. But the political winds aren’t exactly blowing toward aggressive tech regulation right now. The article notes the Trump administration’s pro-AI stance, including a December executive order aimed at limiting state-level AI rules. An FTC investigation would be a major fight. I think Meta calculated the risk and decided the revenue potential from hyper-personalized ads was worth it. Unless regulators step in or users revolt en masse, this is just the new normal. Basically, read the privacy policy because your AI small talk isn’t so small anymore.
