From content moderation to interface design
The European Commission has crossed a new line in platform regulation. On 10 July 2026, it issued preliminary findings that Instagram and Facebook's "addictive design" — infinite scroll, video autoplay, push notifications, and engagement-optimised recommender feeds — breaches the Digital Services Act (DSA) (European Commission, 10 July 2026). This is not a dispute about what content Meta hosts or how it moderates hate speech. It is a dispute about the shape of the product itself — the mechanics that keep a scroll going.
The finding closes out an investigation opened in May 2024, when the Commission first suspected Meta of infringing DSA Articles 28 (minor protection), 34 (risk assessment) and 35 (risk mitigation) (TechCrunch, 16 May 2024). Two years on, the Commission says its evidence shows Meta "did not adequately assess the risks of its addictive design on the physical and mental wellbeing of users, including minors and vulnerable adults," and that Meta's existing mitigation measures failed to address those risks (European Commission, 10 July 2026). Executive Vice-President Henna Virkkunen framed it as a health mandate: "protecting the physical and mental health of Europeans must be a priority for social media platforms."
The steelman: this is what Article 34 was built for
The Commission's legal theory is not a stretch of the DSA — it is closer to the statute's core design. Article 34 requires "very large online platforms" (VLOPs) to identify and assess systemic risks their services pose, including to users' physical and mental wellbeing; Article 35 requires proportionate mitigation (European Commission, DSA VLOP obligations). Lawmakers wrote those provisions precisely because engagement-maximising design — not just illegal content — was understood to be a source of harm, especially to minors. There is a real and growing evidence base, cited repeatedly by EU officials, that infinite scroll and autoplay are engineered to suppress natural stopping cues, and that recommender systems calibrated purely for watch-time will, by construction, favor the most compulsive content over the healthiest. A regulator that only ever polices what is posted while ignoring how the interface behaves is regulating half the problem. On that view, this case is not scope creep — it's the DSA doing exactly what it was designed to do to the largest platforms operating in Europe.
Where the proportionality problem starts
The steelman holds up until the remedy stage, and that is where this case gets harder to defend. The Commission isn't just asking Meta to prove it assessed the risk — it is prescribing specific product defaults: disable autoplay and infinite scroll, and make the recommender "less engagement-oriented" (Al Jazeera, 10 July 2026). That is a regulator co-designing a product's UX, not auditing a risk process. Infinite scroll and autoplay are not unique to Meta or even to Big Tech — they are near-universal norms across video, news, and shopping apps precisely because they reduce friction that users themselves often don't want (manually clicking "next" is not a feature anyone asks for). Mandating specific interface mechanics by regulatory fiat, rather than requiring measurable outcome-based risk mitigation, invites the Commission into product-design decisions it has no comparative advantage in making — and sets a template other regulators, with less rigorous evidence standards, may copy more crudely.
Meta disputes the finding on the merits, not just the remedy. A company spokesperson told reporters: "We disagree with these preliminary findings, which don't accurately take into account the significant steps we've taken to protect teens," pointing to Teen Accounts — which cap under-16 daily use, force overnight sleep mode, and default to private settings — as evidence the company has already been mitigating the risk the Commission is citing (Al Jazeera, 10 July 2026). That defense matters for the proportionality question: DSA Article 35 asks for proportionate mitigation, not the elimination of every engagement-optimising feature outright. If Teen Accounts already measurably reduce night-time and compulsive use among minors, the case for Brussels additionally mandating specific default settings across the entire adult user base weakens considerably.
Stakes and what comes next
The findings are preliminary; Meta can review the Commission's file and respond before the European Board for Digital Services is consulted and a final decision issues. If confirmed, the exposure is real: a fine capped at 6% of Meta's global annual turnover, which analysts peg at more than $12 billion based on last year's revenue (European Commission, 10 July 2026). That scale of exposure is itself a policy lever — it pressures Meta toward EU-specific product changes well before any final ruling, the same dynamic already visible in DSA cases against AliExpress, TikTok, and X.
The better long-term path is one the Commission has the tools for already: require platforms to publish independently audited evidence that their recommender and notification systems don't disproportionately harm minors, and let companies choose how to hit that bar — rather than Brussels specifying autoplay defaults feature-by-feature. Proportionate regulation targets outcomes; only the weakest version of it starts dictating scroll mechanics by name.