Could Congress Regulate Compulsive App Design Features? The Constitutional Issues

Last updated 3 weeks ago. Our resources are updated regularly but please keep in mind that links, programs, policies, and contact information do change.

A single button in Fortnite could produce an unwanted purchase, according to the Federal Trade Commission (FTC). A personalized social-media feed, by contrast, can reflect editorial choices protected by the First Amendment under the Supreme Court’s decision in Moody v. NetChoice. The screen may look seamless to a user, but the law sees several different acts inside it.

Congress can regulate compulsive app design, especially when a rule targets deceptive sales practices or precisely defined product mechanics in interstate commerce. It faces a harder constitutional problem when a rule controls which lawful posts appear, pressures a platform to suppress speech, or makes adults prove their age to read it. The decisive question is not whether software counts as speech. It is what conduct a law regulates, whose speech it burdens and how closely the remedy fits the harm.

The same design can sell, steer and speak

The FTC uses “dark patterns” for interfaces that manipulate choices people might not otherwise make. In Bringing Dark Patterns to Light, FTC staff described privacy choices that appear real but nudge people toward sharing more data. The report connected manipulative interfaces to financial harm, lost autonomy and privacy, among other injuries.

That category is broader than an addictive feed. A hidden charge, an endless scroll and a recommendation system may all keep a person clicking, yet each presents a different legal target. The Supreme Court described Facebook and YouTube feeds as personalized collections chosen from billions of items through prioritization algorithms. Regulating the payment button concerns a transaction. Regulating the order of posts can alter a platform’s presentation of other people’s speech.

The public-health evidence also calls for care: an official report available through CDC Stacks said a causal link had not been established, while the 2023 Surgeon General advisory concluded that social media may harm some teenagers. California lawmakers cited studies associating heavier use with poor mental-health outcomes and less healthy sleep. Those findings support concern and further study, but an association does not identify which feature caused an injury in a particular child.

Congress has broad reach, not a blank check

The Supreme Court has said Congress may reach economic activity that, when aggregated, substantially affects interstate commerce. The Court has also rejected the idea that every remote commercial consequence is enough to create federal power. A national app selling ads, subscriptions or digital goods sits much closer to interstate commerce than the noneconomic activity at the edge of that limit.

Federal law already authorizes the FTC to police unfair or deceptive acts affecting commerce. An unfair practice must cause or likely cause substantial consumer injury that people cannot reasonably avoid, and countervailing benefits must not outweigh it. That test asks for evidence of injury, avoidability and benefits. It is a tighter fit for a confusing purchase path than for a sweeping judgment that an entire style of feed is harmful.

The First Amendment line runs through the feature

In Moody v. NetChoice, the Supreme Court held that platforms receive First Amendment protection when they use standards to decide which posts a feed displays and in what order. The Court also stressed that a facial challenge must show a substantial number of unconstitutional applications relative to a law’s legitimate sweep.

It sent the cases back because lower courts had not examined each covered application and disclosure separately. Justice Samuel Alito’s concurrence added an important limit: not every compiler exercises protected editorial discretion, because some services operate as passive receptacles for other people’s speech.

That reasoning cuts against slogans on both sides. A platform cannot make every interface choice immune by calling code editorial. Government also cannot avoid the First Amendment by calling every burden a product-safety rule. Courts inspect the particular feature, application and remedy.

In Free Speech Coalition v. Paxton, the Supreme Court applied intermediate scrutiny to age verification for material obscene to minors. The Court recognized that adults have a right to the speech and that verification burdens that right, while holding that there is no separate right to avoid verification. The holding did not give lawmakers a universal formula for age gates across social media. It addressed a narrower setting in which the underlying material could lawfully be restricted for minors.

Epic shows the firmer ground

The FTC finalized an order settling charges that Epic Games used confusing button configurations that led to unintended purchases and allowed children to incur unauthorized charges without parental involvement. Epic agreed to pay $245 million for refunds, obtain affirmative consent before charging and stop blocking accounts after disputed charges. The order also prohibits Epic from charging through dark patterns or without affirmative consent. By June 25, 2025, the agency had announced $126 million in another round of refunds and reopened its claims process.

The government did not tell Epic which stories to feature or which views to favor. It tied the remedy to billing, consent and access to purchased accounts. That is why deception and unwanted charges offer Congress and the FTC comparatively firm ground.

Protecting children strengthens the case and complicates the cure

A separate Epic settlement involved the Children’s Online Privacy Protection Act Rule, often called COPPA. The federal rule defines a child as someone under 13. Its statute covers child-directed services and operators with actual knowledge that they collect a child’s personal information. Covered operators must give notice and obtain verifiable parental consent. A design law can likewise focus on minors, but the service must first know which users are minors.

In congressional testimony, NetChoice warned that mandatory verification may collect sensitive information, often government identification, from everyone and create high-value data repositories for criminals. The American Civil Liberties Union argues that age verification can deter anonymous browsing and increase the risk of a breach of identifying information.

Supporters answer that voluntary controls have not worked: Joel Thayer told Congress that existing parental protections had failed and called for enforceable duties stated in law.

Senator Jack Reed, a cosponsor of the current Kids Online Safety Act, says the bill would shift the burden from families to platforms through protective defaults, parental tools and greater transparency. California lawmakers cite algorithmic recommendations, infinite scroll, autoplay and notifications as features that mirror behavioral reinforcement systems associated with addiction. Supporters do not argue that every long session proves addiction. They argue that specific engagement mechanics exploit predictable vulnerabilities while families lack enforceable control over how those mechanics operate for children.

State laws are drawing the constitutional map

California’s Protecting Our Kids from Social Media Addiction Act requires parental controls over nighttime notifications and daily access, with a default block from midnight to 6 a.m.. Beginning January 1, 2027, an operator must reasonably determine that a user is not a minor, obtain parental consent, or fit the law’s actual-knowledge path for providing an addictive feed. The Ninth Circuit said the meaning of “reasonably determine” would depend on regulations from the California attorney general.

NetChoice argued that SB 976 was content based because of its commercial-site exception, focus on social media and rule about displaying reaction counts. The Ninth Circuit ordered an injunction against that like-count provision but otherwise affirmed the denial of preliminary relief. The split result is more instructive than a simple win for either side. A court allowed much of a child-protection framework to proceed while stopping a provision that directly controlled expressive display.

California approved Assembly Bill 1709 on September 10, 2026. The new law bars covered platforms from providing addictive features to users under 16. That is a separate rule from the earlier law’s parental-consent path. It takes effect January 1, 2027.

In litigation over California’s online design code, the Ninth Circuit noted that the requirement for a data-protection impact report remained under a preliminary injunction. The same opinion emphasized that a law is impermissibly vague when ordinary people lack fair notice of what conduct it forbids or it invites arbitrary enforcement.

An Arkansas federal court applied that concern in April 2026, preliminarily enjoining Act 900 after an earlier state social-media law had been permanently enjoined on First Amendment and vagueness grounds. Precision matters twice: it tells companies what to change and lets a court see whether the law reaches conduct, speech or both.

The federal proposal targets defaults and duties

The current federal proposal is the Kids Online Safety Act, S. 1748, introduced by Senators Marsha Blackburn and Richard Blumenthal. On August 5, 2026, the Senate Commerce Committee advanced S. 1748 with the Blackburn substitute, as modified. That is committee action, not enactment. The specific mechanics below come from the introduced text, so they should not be read as a description of every change in the committee substitute.

The introduced bill would limit by default features such as infinite scroll, autoplay, time rewards and notifications for known minors. It would also let minors opt out of personalized recommendations while retaining a chronological feed. For users a platform knows are minors, safeguards would start at the most protective setting the platform offers, with accessible tools for parents. Those provisions regulate mechanics and defaults more directly than the viewpoint of any post.

The ACLU responds that a broad duty tied to anything that increases minors’ activity could cause platforms to suppress protected speech to avoid liability. The Electronic Frontier Foundation favors a different pressure point: privacy rules and a ban on behavioral advertising that would reduce incentives to collect and exploit children’s preferences. This disagreement is about both constitutional risk and causal strategy. Supporters would constrain features built to increase use. Critics would attack the data and advertising incentives underneath them, while avoiding duties that may spill into lawful content.

A durable law starts with a precise target

The existing federal unfairness test requires substantial injury that consumers cannot reasonably avoid, balanced against benefits to consumers or competition. The Ninth Circuit’s warning about vagueness requires lawmakers to give ordinary people fair notice of prohibited conduct. Moody requires courts reviewing a broad law to examine the distinct activities and disclosures it covers.

A federal law is most secure when it names the mechanic, documents the injury and matches the remedy to that injury. Rules against unauthorized charges can demand consent without choosing speech. Default time controls can change how a service operates without necessarily changing which ideas it carries. Controls on recommendations, like counts or access for adults require closer attention because they can alter presentation or burden protected reading.

Even a disclosure can fail if it is not purely factual and uncontroversial, reasonably related to a substantial government interest, and neither unjustified nor unduly burdensome. Labels such as “safety,” “design” and “transparency” do not decide the constitutional test. The operative text does.

The unwanted Fortnite purchase and the personalized feed return the question to its cleanest form. Both happen through software, but only one necessarily involves selecting and arranging protected expression. Congress has room to act between those poles. How much room depends on whether lawmakers regulate a provable commercial mechanism with precision or place government pressure on the speech people may create, arrange and read.