In recent weeks, the company has come under scrutiny from European regulators over its compliance with child protection obligations under the Digital Services Act and has also faced criticism over an AI-powered image generation feature on Instagram.

EU Scrutiny Over Children’s Privacy and Safety

On April 29, 2026, the European Commission, in a press release, stated that Meta’s failure to diligently identify, assess and mitigate the risks of minors under 13 years old accessing their services was in breach of the Digital Services Act (DSA). The commission noted that even though Meta has set the minimum age bar as 13 for these platforms, the restrictions are not that effective as children are still able to get access by providing false dates of birth and there is no proper verification of the same.

The Commission also criticized Meta’s mechanism for reporting underage users as cumbersome and ineffective, noting that reported accounts often remain active without appropriate follow-up. It further noted that Meta’s risk assessment was incomplete as it fails to sufficiently account for scientific evidence demonstrating that younger children are particularly vulnerable to online harms.

Accordingly, the Commission remarked that it is important for these platforms to revise their risk assessment methodology to properly evaluate which risks arise on Instagram and Facebook. The idea is to prevent, detect and remove minors under the age of 13 from their service. It has also been mentioned that if the findings of the commission are confirmed, Meta could face fines of up to 6% of its total worldwide annual turnover under the DSA.

Later on July 10, 2026, the commission issued another set of findings under the Digital Services Act, stating that the design of Facebook and Instagram may expose users, particularly minors, to addictive patterns of use through features such as infinite scroll, autoplay, push notifications, and the platforms’ highly personalised recommender systems.

According to the investigation, Meta failed to adequately assess the risks of its addictive design on the physical and mental wellbeing of users, which includes both minors and adults. These features make the users want to scroll more and shifts the brain into ‘autopilot mode’, contributing to unhealthy habits and compulsive use.

The Commission further observed that Meta had not properly evaluated information of its services being used by minors during late night which increases screen time and the impact of formats such as reels and stories. It questioned the effectiveness of Meta’s existing mitigation measures, including time management tools, parental controls and awareness initiatives, stating that they did not adequately reduce the risks associated with the platforms design. It recommended that Meta should introduce design changes, such as disabling key addictive features like ‘autoplay’ and ‘infinite scroll’ by default, implementing effective ‘screen time breaks’, and adapting its recommender system to make it less engagement-oriented. 

Further, on July 13, 2026, European Commission President  Ursula von der Leyen, announced that the Commission would introduce a draft law in the autumn to strengthen age-appropriate restrictions on social media. The announcement followed the publication of a report by an independent expert panel on child safety online, which recommended delaying access to “social media plus” platforms, including social media services with features such as AI chatbots and certain video games, for children under the age of 13. Referring to “predatory algorithms”, von der Leyen said the proposal was aimed not at restricting children’s access to the internet altogether, but at ensuring that online platforms adopt stronger safeguards before they can reach young users. The proposed legislation comes against the backdrop of the Commission’s ongoing investigations into Meta and TikTok over allegedly addictive platform design, signaling a broader shift towards regulating platform architecture and child safety across the European Union.

Instagram’s AI Feature Sparks Fresh Privacy Concerns

In another news, Meta faced backlash over its newly launched AI image generation tool, Muse Image, which allows users to generate AI images by referencing content from public Instagram profiles. The feature enables users to tag public Instagram accounts and generate AI-generated images using the faces and content appearing in those posts, without notifying the account holders whose images are being used. Meta submitted that the accounts that belonged to users users under 18 are automatically excluded and adult users with public accounts can opt out with easy-to-use controls.

Delhi High Court Examines Alleged Abuse of Facebook’s Copyright Enforcement Mechanism

Recently, two content creators Neeraj Joshi and Pushkar Raj Thakur, approached the Delhi High Court alleging misuse of Facebook’s copyright enforcement system. They contended that Facebook’s “Edit Post” feature is being exploited to falsely establish prior ownership over stolen videos. According to them, infringers first upload unrelated content on Facebook and later replace it with copied videos while the original publication date remains unchanged. The allegedly backdated posts are then used to trigger copyright strikes through Meta’s Rights Manager, resulting in the removal of the creators’ original content.

Joshi has alleged that the loophole has become a tool for extortion, claiming that he received WhatsApp messages demanding money in exchange for withdrawing copyright complaints. He says that despite repeatedly bringing the issue to Meta’s notice and furnishing supporting material, his verified Instagram account was permanently disabled. In a separate suit, Thakur has alleged that copies of his educational videos were circulated online and that the same loophole was used to file copyright claims against his own content. He has argued that Meta failed to introduce adequate safeguards to detect manipulated publication dates and prevent abuse of its copyright enforcement mechanism. Both matters are pending before the Delhi High Court.

Publishers Sue Meta Over AI Training

Meta is also facing a class action before the United States District Court for the Southern District of New York, where major publishers Elsevier, Cengage, Hachette Book Group, Macmillan Publishers and McGraw Hill, along with author Scott Turow, have accused the company of large-scale copyright infringement in the development of its Llama generative AI models. The plaintiffs allege that Meta reproduced and distributed millions of copyrighted books, journal articles and other literary works without authorisation to source, develop and train its AI systems. They contend that Meta copied works through torrenting from pirate websites, web scraping and repeated reproduction during the AI training process, while also removing copyright management information from the works to conceal their origin.

According to the complaint, Meta abandoned discussions to obtain licences from publishers and instead relied on pirated data sets to train Llama, thereby depriving authors and publishers of compensation and undermining the existing licensing market. The plaintiffs have sought relief for alleged violations of the Copyright Act and the Digital Millennium Copyright Act, claiming that Meta’s conduct has caused significant economic harm to authors, publishers and the wider literary ecosystem.

Conclusion:

The developments across the European Union, India and the United States indicate that Meta is facing increasing scrutiny on multiple fronts. While European Commission is examining the company’s compliance with child safety obligations and platform design under the Digital Services Act, proceedings before the Delhi High Court concern the functioning of Meta’s copyright enforcement system, and the class action in the United States raises questions regarding the use of copyrighted works for AI training. The outcome of these proceedings may have significant implications for Meta’s compliance obligations and the regulation of digital platforms and artificial intelligence.