Meta's AI models are trained on your posts. But there are ways to limit what you share.
The company's data sharing practices caught people's attention recently after a Facebook notification was sent to users in Europe informing them about an update to the privacy policy as Meta rolls out new generative AI features in the region. On a separate page, Meta's generative AI privacy policy says it uses "information shared on Meta’s Products and services," including "things like posts or photos and their captions" to train its AI models. It does not use private messages for training data.
SEE ALSO:Meta's new AI tools will get you shopping even more on InstagramA Meta spokesperson said the company is "notifying people consistent with our obligations under local privacy laws" which is presumably to comply with GDPR laws in Europe. According to the notification received by UK-based user Philip Bloom, the changes go into effect on June 26, 2024.
Users in the U.S. did not receive a notification, but based on Meta's data sharing policy, it seems to already be in effect. Meta has been deploying generative AI features since September 2023. It started with the ability to tag the Meta AI chatbot in conversations on Messenger, Instagram, and WhatsApp, and chat with "AI personas" which are characters based on licensed celebrity likenesses including Snoop Dogg, Charli D'Amelio and Kendall Jenner.
More recently, Meta expanded its AI-powered features by making Meta AI the default search bar on Facebook, Instagram, Messenger, and WhatsApp and enabling users to chat with the AI about specific posts. Unfortunately for many who found this irksome, you can't turn it off.
UK and EU based users have a "right to object," meaning they can opt out of sharing their data to train Meta's AI models. That said, it's a tedious process that seems deliberately difficult to navigate. "I'm legit shocked by the design of @Meta's new notification informing us they want to use the content we post to train their AI models," said X user Tantacrul who detailed the many steps required. "It's intentionally designed to be highly awkward in order to minimise the number of users who will object to it."
Tweet may have been deleted
The only way to truly cut off your data from Meta is to delete your accounts. But there are a few different methods for limiting how much data you share.
When we tried clicking on the opt-out link we were shown a message saying the option was only available to people in certain regions. Another link discovered by PCWorld sent us to a Meta help center page where you can submit a request to access, delete, or file a complaint about personal information "from third parties being used to develop and improve AI at Meta."
To fill out the form, you're given three options related to third party data being used for "improving AI at Meta":
"Access, download or correct any personal information from third parties used for building and improving AI at Meta"
"Delete any personal information from third parties used for building and improving AI at Meta"
"Submit a concern about my personal information from third parties that’s related to a response I received from an AI at Meta model, feature or experience"
You'll notice that there's no explicit mention of opting out of sharing your data with the models. The options are narrow and relate specifically to third parties. The form also says request aren't automatically fulfilled; Meta will review your request based on local laws. So users in the EU or UK that fall under stricter privacy laws might have an easier time deleting or accessing their data.
These are the only options available for submitting requests related to sharing your personal information with Meta AI.Credit: Screenshot: Mashable / MetaOnce you've chosen the relevant option, Meta then requires you to fill out your country of residence, name, email, and specific evidence explaining why you're submitting your request.
Users are required to submit specific evidence in order to request personal data being removed. Credit: Screenshot: Mashable / MetaThe form asks you to submit "any prompts you entered that generated a response that included your personal information" or upload a screenshot of the response. So essentially, you're required to give a specific example of why you wish to delete your data and even then, Meta might not fulfill the request.
Another way to protect your data from being used by Meta may be via your "activity off Meta" settings. On this page, you can see sites and apps that share information with Meta. You can choose to disconnect specific sites that share data with Meta, clear previous activity, and manage future data sharing.
In this setting you can see third parties that share information with Meta.Credit: Screenshot: Mashable / MetaDisconnecting and clearing previous data activity will delete data already collected by Meta. But to prevent data from being shared with third parties in the future, choose "Manage future activity" and select "Disconnect future activity." So now, you've cleared your activity history and prevented future data from being shared with third parties.
Here you can block Meta from receiving information about you from third parties. Credit: Screenshot: Mashable / MetaThat said, it's unclear how that impacts data used to train Meta's AI models. Meta's privacy settings are largely related to sharing with third parties, which doesn't address how Meta uses your data internally. It's not a perfect solution, but it's a good start to reclaiming your privacy. We've reached out to Meta for further clarification and will update this story if we received a response.
TopicsArtificial IntelligenceMeta
相关文章：
相关推荐：