Jul 18, 2026
Policy

EU presses Meta to switch off addictive design features or face fines

The European Commission says Facebook and Instagram design choices may breach the Digital Services Act, putting Meta at risk of fines up to 6% of global turnover.

Dominic Okoye

By Dominic Okoye · Staff Writer

· 4 min read

EU presses Meta to switch off addictive design features or face fines
Photo: Ars Technica

The European Commission on Thursday escalated its case against Meta, saying its early findings show Facebook and Instagram use design features that may be addictive and harmful to users, including minors and vulnerable adults. If the Commission reaches a final non-compliance decision under the Digital Services Act, Meta could face fines of up to 6% of its global annual turnover, a figure the Commission did not specify in euros.

The Commission pointed to autoplay, endless feeds and highly personalized recommendation systems as features that can keep users scrolling for longer than intended. EU tech chief Henna Virkkunen told Reuters that the regulator’s position is that Meta’s design is too addictive and must change.

Meta will be able to challenge the preliminary findings over the next several months. Ben Walters, a Meta spokesperson, told Reuters the company disagrees with the Commission’s assessment and said it does not reflect steps Meta says it has taken to protect teens.

Walters cited Meta’s Teen Accounts, which the company says apply protections automatically and give parents tools to block Instagram at night and limit daily screen time to 15 minutes. The Commission was not persuaded by the current package. It said Meta’s default time-management tools for teens do not adequately address risks tied to the design of the services, and that parental controls depend on parents having enough technical skill, time and attention to use them effectively.

What Brussels wants changed

The Commission said Meta should consider turning off autoplay and infinite scroll by default, adding stronger screen-time breaks and changing recommendation systems so they are less oriented around engagement. Virkkunen said Meta can either adjust its products or face a non-compliance decision.

The case is another test of how far the Digital Services Act can go beyond content moderation and into product design. For large consumer platforms, the message from Brussels is that engagement mechanics can become a regulatory liability when they are tied to alleged health risks.

The scrutiny is not limited to Europe. The Commission said it expects expert findings on Monday that could support a Europe-wide social media ban for teenagers. Such a move would directly affect platforms that depend on young users, and the Commission cited internal Meta messages showing the company wanted to keep people engaged on its services for life.

In the United States, Meta also failed recently to dismiss a lawsuit brought by 29 states alleging that Facebook and Instagram addict children. Reuters reported that the trial is scheduled to begin in August and that states may seek up to $1.4 trillion in penalties if they prevail. Reuters noted that figure is close to Meta’s roughly $1.5 trillion market capitalization. California Attorney General Rob Bonta told Reuters that Meta put profit ahead of children’s safety.

Regulatory risk meets AI spending

The timing is awkward for Meta because its capital demands are rising. The Information reported that Mark Zuckerberg has allocated between $125 billion and $145 billion in capital expenditures this year for AI data centers, with higher operating expenses tied to outside cloud services and AI talent. The Information also reported that Meta’s Reality Labs unit has lost $87 billion since 2020.

Business Insider reported that Meta told employees at an internal town hall that an upcoming AI model, code-named Watermelon, had caught up with OpenAI’s GPT-5.5. Business Insider also reported that Meta has struggled to persuade developers and customers that its models are at the front of the market.

Meta is also facing criticism over Muse, an AI model that uses public Instagram photos and videos. NBC News reported that its tests found Muse could be used to make deepfakes of celebrities and ordinary people, and that Meta’s detection tools did not catch every case. Meta said most users were opted in by default, except private accounts and users under 18 whose sharing and reuse settings were off by default. Meta told Reuters users can opt out in a few clicks, but it has not explained why default opt-in was necessary.

This story draws on original reporting from Ars Technica.

More from Policy

All Policy →