Meta, the company behind Facebook and Instagram, is set to begin using public posts from European users to train its artificial intelligence (AI) systems starting June 26. This decision has sparked discussions about privacy and GDPR compliance.
Utilising Public Data for AI
European users of Facebook and Instagram have recently been notified that their public posts could be used to help develop Meta's AI technologies. The information that might be utilised includes posts, photos, captions, and messages sent to an AI, but private messages are excluded. Meta has emphasised that only public data from user profiles will be used, and data from users under 18 will not be included.
GDPR Compliance and Legitimate Interest
Under the General Data Protection Regulation (GDPR), companies can process personal data if they demonstrate a legitimate interest. Meta argues that improving AI systems constitutes such an interest. Despite this, users have the right to opt out of having their data used for this purpose by submitting a form through Facebook or Instagram, although these forms are currently unavailable.
Even if users opt out, their data may still be used if they are featured in another user's public posts or images. Meta has provided a four-week notice period before collecting data to comply with privacy regulations.
Regulatory Concerns and Delays
The Irish Data Protection Commission (DPC) intervened following Meta's announcement, resulting in a temporary delay. The DPC requested clarifications from Meta, which the company has addressed. Meta assured that only public data from EU users would be utilized and confirmed that data from minors would not be included.
Meta’s AI Development Efforts
Meta is heavily investing in AI research and development. The company’s latest large language model, Llama 3, released in April, powers its Meta AI assistant, though it is not yet available in Europe. Meta has previously used public posts to train its AI assistant but did not include this data in training the Llama 2 model.
In addition to developing AI software, Meta is also working on the hardware needed for AI operations, introducing custom-made chips last month.
Meta's initiative to use public posts for AI training highlights the ongoing balance between innovation and privacy. While an opt-out option is provided, its current unavailability and the potential use of data from non-consenting users underscore the complexities of data privacy.
European users should remain informed about their rights under GDPR and utilize the opt-out process when available. Despite some limitations, Meta's efforts to notify users and offer an opt-out reflect a step towards balancing technological advancement with privacy concerns.
This development represents a striking move in Meta's AI journey and accentuates the critical role of transparency and regulatory oversight in handling personal data responsibly.
