Meta Platforms is widening the reach of data it receives from outside businesses. What once fed only targeted advertisements will now help decide what appears in users’ Facebook and Instagram feeds. The same information will shape answers from the company’s AI chatbot.
The change takes effect next month. It applies in 10 countries. The Hacker News first detailed the scope on Tuesday. Meta already gathers signals such as games played or items bought on partner sites. Those details have sharpened ad relevance for years. Now the company plans to fold them into recommendation engines that power organic content and AI output.
“Businesses often share information about people’s activity on their sites with us to make ads more relevant,” Meta stated in its announcement. “In the future, we’ll use this information to personalize other parts of your experience, including the content you see in your Feed and AI responses.” The company framed the move as an improvement for users who want material that matches their habits. Yet the expansion blurs lines that privacy advocates have long tried to draw.
This isn’t fresh data collection. Meta insists no new categories are being added. The difference lies in application. Purchases at an online retailer or progress in a mobile game will now inform which Reels autoplay or which topics the AI emphasizes. The result could feel eerily on point. Or intrusive. Depends whom you ask.
Users outside the initial 10 countries face a manual step if they want to block the practice. They must locate the “Activity from other businesses” setting and switch it off. The control sits inside the Accounts Center. One toggle covers ads, feeds, and AI replies together. Simplicity has its trade-offs. A single choice now governs multiple surfaces.
The announcement arrives months after Meta began using conversations with its generative AI tools to refine recommendations. That shift, reported last fall by CNET, drew criticism for turning private chats into behavioral fuel. Critics saw a pattern. First AI interactions. Now off-site commerce. Each step pulls more of daily life into the company’s personalization loop.
Industry watchers note the business logic. Meta’s ad revenue still dominates. Yet competition from TikTok and emerging platforms pressures the firm to lift time spent inside its apps. Better recommendations keep users scrolling. AI responses that reflect recent purchases or hobbies feel conversational. They also keep people engaged with Meta’s chatbot rather than rivals.
But. The move revives old questions about consent. Many users never consciously linked their retail activity to social feeds. They clicked “accept” on some distant cookie banner or downloaded an app that quietly shared purchase data. Those signals, once confined to ad auctions, now influence what friends see shared or what the AI suggests next. The feedback loop tightens.
Meta offers ways to review and clear past activity. The company has maintained an Off-Facebook Activity tool for years. It lets users see which partners sent data and disconnect future flows. Effectiveness varies. Some partners continue sending hashed identifiers that Meta can still match. Complete separation remains difficult without abandoning entire categories of online services.
Privacy advocates at the Electronic Frontier Foundation have argued for stronger defaults. In a January 2025 analysis, the EFF urged users to limit data sharing wherever possible. “Meta tracks your activity across millions of websites and apps,” the organization wrote. The group highlighted how the company’s pixel and partner integrations create persistent profiles even for non-users.
Regulators in Europe have forced some concessions in recent years. Meta now lets EU users opt out of certain personalization tied to on-platform behavior. The current expansion appears limited to markets without those strict rules. The 10 countries were not named in initial coverage but likely exclude the EU bloc. That geographic split echoes past Meta strategies to contain compliance costs.
Analysts expect limited user backlash. Most people never adjust privacy settings. Feed quality improvements could even win approval from those who value relevance over data boundaries. Early reactions on X mixed skepticism with resignation. One technology commentator posted that the line between social media and surveillance keeps getting thinner. Another noted the feeds could soon feel “even creepier.”
Meta’s blog post emphasizes user benefit. The company claims people will see less irrelevant content. AI answers will better match interests inferred from real-world activity. Yet the same system that predicts which sneakers to advertise can now predict which hiking videos to promote or which travel queries to prioritize in chat. The distinction between commercial and editorial personalization fades.
Advertisers stand to gain too. More precise signals across surfaces let them reach audiences through content rather than just sponsored posts. A user who buys running shoes on one site might encounter running communities, gear reviews, and race announcements in their feed. The AI could reference those shoes in conversation. Every interaction reinforces the profile.
Longer term, the strategy aligns with Meta’s heavy investment in AI. The company has poured resources into recommendation models that already dominate feed ranking. Adding off-platform signals gives those models richer training data without additional collection expense. Efficiency matters when quarterly earnings reports highlight AI progress as a growth driver.
Still, trust issues linger. Past scandals involving Cambridge Analytica, shadow profiles, and misplaced user data have left many wary. Each expansion of data use, even when technically permitted under existing policies, prompts fresh scrutiny. Meta’s privacy policy already discloses third-party data for ad personalization. The latest update simply extends that language to feeds and AI. Transparency exists. Comprehension does not always follow.
Users who want to act have options. Visit the Accounts Center. Find the activity setting. Turn it off. Clear previous data if desired. The process takes minutes. Impact appears immediately in reduced personalization. But complete escape from Meta’s data web requires broader steps: browser extensions that block the pixel, VPNs that obscure location, avoidance of partner apps. Few follow that path.
The announcement, covered Tuesday by both The Verge and The Hacker News, signals another incremental shift. No single change feels dramatic. Taken together over years, they reshape expectations about privacy. What happens on other websites no longer stays there. It follows users into their social experience and their conversations with machines.
Meta shows no sign of reversing course. Engagement metrics and ad performance remain paramount. Personalization remains the proven path to both. The company will likely continue refining controls and issuing explanations. Users, for their part, will keep scrolling. The feeds will simply know them a little better.
WebProNews is an iEntry Publication