If you've been using Meta AI in Facebook, Instagram, WhatsApp, or Messenger, you might want to stop. Starting later this year, Meta plans to use your generative AI conversations to personalize content and tailor ad recommendations.
Text exchanges and voice conversations with Meta AI will be used to personalize posts, reels, ads, and more. Meta says that incorporating AI conversations will help the company improve its recommendations so that people are "more likely to see content they're actually interested in."
Meta likens using AI conversations to tailoring recommendations based on the content that people interact with on Facebook and Instagram. If you converse with AI about hiking, for example, Meta will show ads for hiking boots, as it would if you liked a hiking-related page.
There is no way to opt out of Meta's AI-based ad targeting, but Meta claims that users can adjust the content and ads they're seeing with Ad Preferences and feed controls. With Ad Preferences, you can limit ads from specific topics or block advertisers one by one, but you can't choose not to let Meta access your AI conversations. Interactions with Meta AI are used across Meta products that have been added to the same Accounts Center.
With no opt-out feature, the only way to prevent Meta from using your AI conversations for content and ad targeting purposes is not to use the AI in the first place.
Meta says that sensitive topics like sexual orientation, political views, health, racial or ethnic origin, philosophical beliefs, or trade union membership are not used for ad targeting purposes. Conversations that take place before December 16 won't be used.
Meta plans to start using AI for personalization on December 16, 2025, and customers will receive in-product notifications and emails before it happens. Data will be used for personalization in most countries around the world, though not in the EU, UK, and South Korea for the time being.
