Meta

CJEU: Meta must restrict the use of personal data for targeted ads

Today, the Court of Justice of the European Union (CJEU) delivered a landmark ruling restricting the use of personal data for targeted advertising by platforms like Meta, formerly Facebook. That decision stems from a complaint by data privacy advocate Max Schrems, who argued Meta violated the General Data Protection Regulation (GDPR) principles of ‘data minimisation’ and ‘purpose limitation’ when processing users’ data for personalised ads. Consequently, the court has mandated time limits on data usage and stipulated that data must only be utilised for expressly defined purposes under GDPR, impacting the vast data reservoirs companies have accumulated over time.

An essential aspect of the ruling addresses the use of sensitive data, such as publicly shared sexual orientation information, for advertising. Schrems challenged Meta’s adherence to the ‘purpose limitation’ principle when the company used such public data without explicit consent. The court decided that declaring such information publicly does not permit its use for targeted advertising, establishing that public availability does not equate to consent under GDPR. This sets a significant precedent for sensitive data processing and aligns with the broader GDPR framework.

Why does it matter?

The ruling has implications beyond social media advertising, particularly AI training practices. The European Center for Digital Rights (Noyb), founded by Schrems, emphasised that companies like Meta, X (formerly Twitter), and OpenAI often scrape online data to train AI models, frequently diverging from the data’s original intent.

This ruling also highlights the evolving regulatory landscape and the EU’s global data privacy governance leadership. Schrems’ advocacy, informed by past victories in the Schrems I and Schrems II cases, continues to shape strict data privacy standards within Europe, reaffirming the ongoing scrutiny of technology firms’ data practices.

Meta battles scam ads in Australia

Meta and Australian banks have worked together to remove 8,000 fraudulent ‘celeb bait’ advertisements from Facebook and Instagram. The scams, often using AI-generated images of celebrities, deceive users into investing in fake schemes. Australian banks flagged 102 such cases since April.

The rise in these scams has led Australia to draft a new anti-scam law, which could impose fines of up to A$50 million on companies that fail to combat online fraud. Reports in 2023 show that Australians lost a staggering A$2.7 billion to various scams.

Meta is currently facing legal challenges in Australia, including a lawsuit for allowing cryptocurrency ads featuring celebrities like Russell Crowe and Nicole Kidman. Despite these issues, Meta continues its efforts to fight fraudulent ads.

Meta, alongside Australian banks, believes that early signs within ads could help detect wider scam activity. The company is reviewing Australia’s draft legislation, signalling a continued focus on anti-scam measures in the future.

Vietnam to host Meta’s new AI and tech projects

Meta is set to expand its presence in Vietnam by increasing investment in AI and manufacturing. Starting in 2025, production of its latest mixed reality headset, the Quest 3S, will move to the country.

The announcement came shortly after Meta’s president for global affairs, Nick Clegg, met with Vietnamese officials. Vietnam is a key market for Meta, with millions of users relying on its social media platform, Facebook.

Meta also supports numerous small businesses in the region. While the size of the investment remains unclear, Meta’s commitment highlights the growing importance of the Vietnamese market for the company.

Additionally, Meta plans to further integrate AI innovations into its platforms. Testing of its Meta AI tool in Vietnamese will begin soon, with a full launch expected by the end of the year.

Germany classifies Microsoft as major competitor

Germany’s competition watchdog has designated Microsoft as a ‘company with paramount cross-market significance for competition,’ allowing for stronger regulatory actions against the tech giant. Andreas Mundt, head of the Bundeskartellamt, emphasised that Microsoft’s products are essential across various sectors, making its ecosystem more interconnected than ever.

This classification also extends to Apple, Google, and Meta, indicating that Microsoft will undergo heightened scrutiny and may face restrictions on anti-competitive practices. A spokesperson for Microsoft expressed the company’s commitment to promoting a competitive environment and working cooperatively with the Bundeskartellamt.

The designation follows antitrust charges against Microsoft by the European Commission in June, which accused the company of unfairly bundling its Teams app with the Office suite, putting rivals like Slack at a disadvantage. The German authority clarified that its increased oversight will apply to Microsoft as a whole, rather than focusing solely on individual services or products.

Apple faces limited claims in data privacy case

A federal judge has scaled down a privacy lawsuit against Apple, which alleged the company collected personal data from iPhone, iPad, and Apple Watch users without permission. The lawsuit targets Apple’s apps, including the App Store, Apple Music, and Apple TV. US District Judge Edward Davila dismissed most claims involving the “Allow Apps to Request to Track” setting, clarifying that it only governs data collection by third-party apps and websites, not Apple’s in-house apps.

Despite dismissing many claims, the judge allowed some to proceed related to Apple’s ‘Share [Device] Analytics’ setting. The plaintiffs claim that Apple continued collecting data even after users disabled the setting, despite promises that it would stop data sharing. Judge Davila agreed, noting that users could reasonably assume they had withdrawn consent based on Apple’s own disclosure that disabling the option would prevent data collection.

This lawsuit is part of a broader trend of legal actions against major tech companies like Google and Meta, accusing them of gathering user data without proper consent. Neither Apple nor the plaintiffs’ lawyers have responded to requests for comment on the case as it unfolds.

EU hits Meta with €91 million fine for password security breach

Meta, Facebook’s owner, has been fined €91 million ($101.5 million) by the EU’s privacy regulator for mishandling user passwords. The issue, which surfaced five years ago, involved Meta storing certain users’ passwords in plaintext, a format lacking encryption or security protection. Ireland’s Data Protection Commission (DPC), which oversees GDPR compliance for many US tech firms operating in the EU, launched an investigation after Meta reported the incident.

Meta admitted the error, emphasising that third parties had not accessed the exposed passwords. However, storing passwords in an unprotected format is considered a major security flaw, as it exposes users to significant risks if unauthorised individuals access the data. Deputy Commissioner Graham Doyle underscored that storing passwords without encryption is widely unacceptable due to potential abuse.

This fine adds to Meta’s growing list of penalties under the EU’s General Data Protection Regulation (GDPR). To date, Meta has been fined a total of 2.5 billion euros for various data breaches, including a record €1.2 billion fine in 2023, which Meta is currently appealing. These repeated infractions highlight ongoing concerns about how the company handles sensitive user data.

Meta postpones joining EU AI Pact, focuses on compliance

Meta Platforms has announced it will not immediately join the European Union‘s voluntary AI Pact, which is a temporary initiative ahead of the AI Act coming into force. The company is currently focusing on compliance with the forthcoming regulations set out in the act, but may sign the pact at a later stage.

The EU’s AI Act, agreed in May and adopted by the European Council, will introduce strict rules governing the development and use of artificial intelligence. Under these regulations, companies must provide detailed summaries of the data used to train their AI models. The majority of the law’s provisions will take effect from August 2026.

In the interim, the AI Pact encourages companies to voluntarily adopt some of the key requirements of the forthcoming act. Meta has expressed its support for harmonised EU regulations but is prioritising work on meeting the obligations of the AI Act.

The AI Act will be part of a wider legislative framework, joining the Digital Markets Act, Digital Services Act, Data Governance Act, and Data Act, in shaping the future of digital regulation in the EU.

Meta introduces prototype of Orion AR glasses

At its annual Connect conference, Meta Platforms unveiled its first working prototype of augmented-reality glasses called Orion. CEO Mark Zuckerberg described the chunky black glasses as a glimpse into a future where virtual and physical worlds merge seamlessly, referring to them as a “time machine” that could transform user interactions. The announcement also featured improved AI chatbot capabilities and a new Quest mixed-reality headset, contributing to a record closing high for Meta shares at $568.31.

The Orion glasses, made from magnesium alloy and powered by custom silicon designed by Meta, will include features like hand-tracking, voice controls, and a wrist-based neural interface. Meta plans to refine the glasses to make them smaller and more affordable for a projected consumer launch in 2027. However, previous attempts at AR by major tech companies have often encountered challenges. Analysts recognise Meta’s goal of making augmented reality accessible, but public scepticism about AI technology continues to be a significant barrier.

Although Zuckerberg did not demonstrate the glasses’ features live, a video showcased testers, including Nvidia CEO Jensen Huang, interacting with the device. Meta’s existing Ray-Ban smart glasses gained popularity after the introduction of an AI assistant, which will soon allow users to scan QR codes and stream music using voice commands. Future updates for these glasses are set to include real-time language translation and video generation capabilities.

Alongside its AR announcements, Meta unveiled several AI updates, including improved audio responses for its digital assistant, Meta AI, which can now mimic celebrity voices. With over 400 million monthly users, Meta is heavily investing in AI and AR technologies, anticipating record capital expenses of $37 billion to $40 billion for 2024. However, despite these investments, the Reality Labs division reported substantial losses of $8.3 billion in the first half of this year.

Zuckerberg highlights Meta AI’s expansion to 500 million users

Meta AI is fast becoming one of the world’s most widely used assistants, with nearly 500 million monthly active users, according to Mark Zuckerberg. Speaking at Meta Connect 2024, he highlighted its growth, pointing out that major markets like the EU have not yet been tapped.

India remains the largest user base for Meta AI, thanks to WhatsApp’s popularity, which boasts over 500 million users in the country. The AI is rapidly expanding its reach, placing it far ahead of competitors like OpenAI’s ChatGPT, which has around 200 million weekly users.

Alongside usage updates, Meta revealed new developments, including the Llama 3.2 models, offering enhanced multimodal capabilities. These models will be available in Europe, though they do not rely on European user data, due to regulatory concerns.

The event also introduced celebrity voices and new AI-powered features across platforms like Facebook, Instagram, and Messenger. Meta AI can now create photos, understand user images, provide answers, and even make edits.

Celebrities fall for ‘Goodbye Meta AI’ hoax

Over 600,000 users, including celebrities like James McAvoy, Ashley Tisdale, and Tom Brady, have been tricked by a viral hoax on Instagram. The message claimed sharing the ‘Goodbye Meta AI’ post would prevent Meta, owner of Facebook and Instagram, from using personal images for AI training.

The viral trend led many to believe posting the message would revoke Meta’s ability to use their data. However, fact-checkers, including Lead Stories, have labelled these posts as false, confirming that users can only opt-out via account settings, not by posting stories.

Meta confirmed that the viral post, which gained momentum after major celebrity shares, has no impact on privacy settings. The hoax stemmed from opposition to Meta’s June announcement regarding the use of public posts to train its AI models.

Though the trend began on 1 September, its popularity spiked after celebrities shared it, with Google Trends showing increased interest after 24 September. This instance echoes past privacy-related hoaxes and reflects public unease about AI data usage.