Privacy group complains to UK regulator about Meta scraping user data to train AI

A UK data rights campaign group has launched a complaint with the data law regulator against Meta’s change of privacy policy which allows it to scrape user data to develop AI models.

Open Rights Group (ORG), a UK-based membership organization that campaigns to protect privacy and free speech online, pointed out that Meta emailed Facebook and Instagram users in the UK at the end of May to warn it would introduce changes to its privacy policy on June 26. According to the ORG, Meta would “rely on the legal basis called legitimate interests” to use individuals’ information for its AI development.

The complaint follows a similar protest in the European Union under the General Data Protection Regulation, which resulted in the social media biz agreeing to pause plans to train AI models on EU users' Facebook and Instagram users' posts.

Although UK data protection law currently mirrors that in the EU, its decision to leave the economic and political bloc came into effect at the end of 2020.

The ORG pointed out there had been no official change to the Meta privacy policy to make the ending of data processing for the development of Meta's "AI technologies" legally binding in the UK. It has therefore raised a formal regulatory complaint with the nation’s data protection regulator, the Information Commissioner’s Office (ICO).

Mariano delli Santi, complainant and Legal and Policy Officer at Open Rights Group, said: “Meta’s plans to ingest its users’ data, posts and pictures will impact more than 50 million Instagram and Facebook users in the UK. It’s not acceptable that the company is making a half-hearted attempt to enable people to opt out rather than give their consent to such intrusive data processing.

“The proposals appear to violate UK GDPR on a number of levels, and we urge the ICO to investigate thoroughly and stop them once and for all,” he said. The ORG added that while Meta told users they had the right to object, it did not commit to honoring objections as a matter of course and once a user’s data had been used by the company, any consent could not be applied retrospectively.

The ICO has yet to respond to The Register's request for comment.

In June, Meta agreed to pause its plans to train AI models on EU users' Facebook and Instagram posts following pressure from EU regulators. It said the move would delay its plans to launch Meta AI in the economic zone.

In a blog following the move, Stefano Fratta, global engagement director, Meta Privacy Policy said: "We remain highly confident that our approach complies with European laws and regulations. AI training is not unique to our services, and we’re more transparent than many of our industry counterparts."