Meta is currently training its generative AI models. As it stands, both users and non-users of its products should be worried about how their personal data is being used to train the AI models. The tech company relies on user generated content as part of its training data set.
The company’s generative AI privacy policy says, “We also use information shared on Meta’s Products and services. This information could be things like posts or photos and their captions. We do not use the content of your private messages with friends and family to train our AIs.”
Privacy remains a big concern. Meta concedes that it gathers data from the internet or licenses data from other providers to train it models. These data may include personal information including usernames, home addresses, phone numbers and email addresses.
Meta claims, “If we collect a public blog post it may include the author’s name and contact information. When we do get personal information as part of this public and licensed data that we use to train our models, we don’t specifically link this data to any Meta account.”
Alarmingly, this data collection is not only limited to those who use Meta products or services. Meta says information from everyone can still find way into its AI models.
“This could happen if you appear anywhere in an image shared on our Products or services by someone who does use them or if someone mentions information about you in posts or captions that they share on our Products and services,” the tech giant states.
Further, the company states on its website that information from third parties includes what is publicly available on the internet and licensed information that someone else holds the rights to and gives Meta permission to use.
Meta AI Opt Out Request
Currently, the company has provided opt-out forms. However, many users don’t know about this. Secondly, with information from third parties being used, it’s unclear how effective opting out is in reality.
READ: Netflix/Facebook Exchanged User Private Messages, Watch History in Scandalous Data Breach
This objection form requires users to share reasons with Meta as to why don’t want their data to be used. Meta has no obligation to honour the request. Navigating the opt-out process for their generative AI models is not an easy task.
On Facebook users need to click their profile picture in the top right corner. Next is to select “Settings & Privacy” followed by “Privacy Center.” This is followed by scrolling down and locating the obscure “Generative AI at Meta” section.
After more scrolling past numerous links, users find a subsection titled “More Resources.” .It is within “More Resources,” that the first link takes you where you woulld like to go. The section is “How Meta uses information for Generative AI models.” But beware, this is a lengthy read – over 1,100 words – before finally revealing a separate link to the elusive “right to object” form.
EU and UK Exemption
Since enacting the General Data Protection Regulation (GDPR), a legal framework that establishes guidelines for how personal information can be collected and processed, the European Union has fought big tech companies on matters data privacy.
This week Meta’s lead regulator in the EU, the Irish Data Protection Commission (DPC), announced that Meta is suspending plans to start training its AI systems using data from its users in the European Union (EU) and U.K. Acting on behalf of several data protection authorities across the EU has been engaging Meta on the matter. In the engagements DPC presented aspects of the GDPR that would be flaunted. The U.K.’s Information Commissioner’s Office (ICO) also requested that Meta pause its plans. The suspension is until Meta could satisfy concerns ICO had raised.
However, Meta recently informed users about an upcoming privacy policy change. The company claims the new policy grants them the right to utilize public content across Facebook and Instagram to train their AI models. This includes content from comments, interactions with businesses, status updates, photos, and their captions. Meta argues this is necessary to ‘reflect the diverse languages, geography, and cultural references of people in Europe.
The policy changes were due to come into effect on June 26. Non profit, NOYB (“none of your business”) filled 11 complaints in various EU countries to bar this. NOYB argue that Meta’s new policy could violate the GDPR in several ways. A key concern is the opt-in versus opt-out approach to user consent. Under GDPR, when processing personal data, companies should obtain explicit permission from users before using the data, rather than forcing them to take action to opt-out.
In other parts of the world including Africa, Meta is already relying on user-generated content to train its AI.
