Meta AI — privacy & data policy

Provider: Meta. PromptPrivacy score: 39/100. Last verified November 01, 2026.

What happens to your data?

Meta collects the content, messages, and metadata generated when interacting with Meta AI. This data is used to provide and personalize features, deliver targeted advertisements, and train Meta's artificial intelligence models. Information is shared across Meta Companies, as well as with external service providers, partners, researchers, and law enforcement when required.

Does it train on your data?

Trains on your content by default. Meta processes user interactions with AI features, including content and messages exchanged with these features, to develop and improve AI at Meta for Meta Products and third parties. The policy does not specify an opt-out mechanism for AI training.

Can you opt out?

Not specified. The policy states that user interactions with AI features are used to develop and improve AI at Meta, but it does not specify any option or mechanism to opt out of AI model training.

Data retention

Meta retains information for as long as needed to provide products, comply with legal obligations, or protect legitimate interests, determined on a case-by-case basis. Extended retention may apply for legal and safety reasons, with no fixed retention period specified for Meta AI interactions.

Encryption & security

Not specified. The policy mentions general efforts to promote safety, integrity, and security, but does not disclose specific encryption protocols such as TLS or encryption at rest, nor does it list security certifications.

Is it safe for work?

No. Content and messages sent to Meta AI are processed by default to develop and improve AI models and may be subject to manual human review. Because Meta also shares data across its family of companies and third parties without enterprise confidentiality guarantees, it is not safe for proprietary or sensitive workplace use.

Source policy: https://www.facebook.com/privacy/policy/