For nearly two decades, the companies operating popular social media platforms remained legally “untouchable” when it came to the negative consequences caused by users’ exposure to their products.

The effects on mental health, the facilitation — and, at times, the live streaming — of criminal acts, including sexual abuse, and other similar issues had triggered strong public backlash, but had never led to lawsuits with a substantial outcome.

About four years ago, however, according to The New Yorker, prosecutors in the United States began turning their attention to the business choices and design decisions of technology companies rather than, as had happened in the past, to the promotion of specific harmful content, since that approach faced significant obstacles because of protections for freedom of speech.

In a case heard during the spring of this year, a young woman who argued that her mental health had suffered a devastating deterioration as a result of using social media was awarded $3 million in damages.

At the same time, in March and August, in a case filed by the Attorney General of New Mexico, a court imposed fines of nearly $1 billion on Meta — the parent company of Facebook and Instagram — and ordered it to make a series of changes to its products and business practices. Meta has already appealed the decision.

In late 2022, amid a wave of lawsuits, cases filed by the attorneys general of 29 states were consolidated into one major class-action case. The allegation was that Meta had knowingly caused a series of harms, primarily because its products “exploit and manipulate its most vulnerable consumers: teenagers and children.”

The case went to trial in Oakland in early August and was expected to last several weeks. However, it was abruptly terminated last Wednesday.

The whistleblower’s testimony paved the way

After testimony from Meta’s former executive and whistleblower Arturo Béhar, who had served as director of engineering for product safety issues and argued that the company culture “made it practically impossible to implement features that addressed the well-being and safety issues we were talking about,” as well as testimony from Instagram head Adam Mosseri, who rejected the lawsuit’s framework characterizing excessive social media use as “clinical addiction,” the company — whose market capitalization stands at $1.5 trillion — agreed to a settlement.

The agreement with 48 states — ultimately — and 52 attorneys general provides for the payment of at least $12 billion over a decade (far less than the $200 billion initially estimated and the $1.4 trillion that the company itself had been floating before the trial) and a commitment to a series of changes to its platforms in the United States market.

The comparison with the Big Tobacco Settlement

Since the agreement was announced, many have compared the development to the 1998 Big Tobacco Master Settlement, when four of the largest tobacco companies agreed to pay enormous sums, accept restrictions on marketing, and provide significant public-health information after acknowledging that they knew their products were addictive and harmful to human health.

The comparison has some basis, even if the two cases are not entirely alike. The agreement reached this week will neither be the final answer regarding the relationship between social media and health nor a panacea that will “cure” all of the problems. Nevertheless, there are serious reasons to consider it a potential turning point.

Although it refuses to legally acknowledge liability, Meta is implicitly accepting, for the first time, that a significant part of its business model is unsuitable for children, their health, and their safety.

The changes Meta agreed to — What the settlement’s 35 pages provide

Meta had spent much of the past two years promoting Teen Accounts, a new set of features it introduced in 2024 that included enhanced privacy options, stricter restrictions on messaging, and a number of other changes.

With the new agreement, however, Meta has essentially acknowledged that these features were not sufficient. To a large extent, these measures shifted the responsibility for supervision to parents rather than to the company itself. The agreement now indicates that this approach is not considered sufficient.

The 35 pages of technical terms and obligations contained in the settlement, which The New Yorker reviewed, provide for new restrictions on Meta’s products, including:

• hiding the number of “likes” on posts, as it has been linked to harmful social-comparison trends,
• new safeguards for parental supervision,
• as well as stronger commitments to ensure that abusive content flagged by users will be addressed within six hours in at least 90% of cases.

Although this latter obligation may seem like a minor technical detail, the evidence shows that for years Meta’s response to reports of problematic content was so limited that many users stopped reporting such material. According to an internal company document from 2022, “the current reporting process on Facebook is extremely cumbersome, almost impossible for users on mobile devices.”

Two-Hour Limits and a “Block” at Night

The most significant concession by Meta under the agreement concerns strict restrictions on teenagers’ use of its platforms.

The company agreed to:

a maximum daily usage limit of two hours,
a complete block on access to content feeds from midnight until 6:00 a.m.,
the elimination of push notifications both during the night and during school hours.

According to the original complaint, these notifications were something Meta itself knew could contribute to teenagers’ “inattention and hyperactivity,” with significant negative effects on their well-being.

The available data clearly demonstrate the scale of young people’s use. According to the U.S. think tank Pew Research, in 2024, 60% of teenagers said they used Instagram, while 12% reported that they opened it “almost constantly.” Other internal Meta documents showed that in recent years, approximately 1% of users spent more than five hours a day using the company’s services.

Until now, users with extremely high levels of engagement generated a large share of advertising revenue. According to analysts, limiting the overall number of these users could strengthen the company’s financial incentives to place greater emphasis on the average user’s overall experience rather than on extreme usage.

The Independent Monitor Who Will Oversee Meta

One of the most important elements of the agreement lies not only in the new restrictions, but also in the provision for continuous oversight by an independent third-party monitor. There is precedent for similar arrangements in the past: In 2019, Meta accepted oversight as part of a settlement with the U.S. Federal Trade Commission (FTC) over privacy violations.

However, this is the first time a technology company has agreed to the appointment of an independent monitor whose sole purpose is the protection of children. Just weeks earlier, in the ruling on the New Mexico case, Meta had opposed the idea of an external monitor, and the judge had declined to impose such a mechanism.

The monitor, who will be jointly appointed by the states and Meta, will be responsible for assessing the company’s compliance with the terms of the agreement.

The monitor will not have the authority to directly impose changes to the company’s conduct. However, under the terms of the settlement, for the next decade they will have access to data, documents, and internal communications comparable to those that would be examined during legal proceedings. At the same time, the monitor will be required to notify the states that participated in the case of any failure by Meta to comply with its commitments, and those states, in turn, may initiate new legal action.

The Gaps in the Agreement: Algorithms and Endless Scrolling

One area in which Meta can claim a “small victory” concerns the way content is selected and displayed on its platforms. Indeed, the way Instagram and Facebook decide what is shown to users — the design of their algorithms and user interfaces — was at the heart of the allegations from the outset that these products are unsafe.

However, the agreement reached does not include some of the most important design features of the platforms, such as infinite scrolling and autoplay.

Former Meta employee and whistleblower Arturo Béhar testified that, according to research he conducted while working at the company, more than half of Instagram users had negative experiences on the platform over a seven-day period. Among them, 24% of teenagers aged 13 to 15 reported having received unwanted sexual advances.

In these cases, the issue is not only the amount of time a user spends on the app, but also the way the app itself creates connections, contacts, and opportunities for communication between people. In this context, Meta CEO Mark Zuckerberg himself acknowledged a few months ago that only 10% of the content appearing on Instagram comes from users’ friends. The rest consists of external content recommended by the platform’s algorithms.

Beyond the content appearing in users’ feeds, recommendation algorithms also influence the accounts teenagers connect with. According to the states’ allegations, these recommended accounts are a key element in the way platforms can facilitate the promotion of sexual abuse.

However, the states’ agreement with Meta appears to largely accept the company’s existing definition of what constitutes “inappropriate” content. Thus, Meta continues to retain significant control over what is classified as a problem and what is not.

Meaningful improvement in these areas may therefore depend on a difficult task for the independent monitor: applying pressure to strengthen the safety standards themselves.

The Big Challenge of Age Verification

For similar reasons, another controversial aspect of the agreement concerns a feature that lies at the foundation of the entire effort: verifying users’ ages.

Meta’s age-verification systems have been at the center of numerous complaints concerning child protection. The agreement calls on the company to devote significant resources to improving these mechanisms. However, while it introduces new standards and requirements, it also gives Meta significant control over assessing how effectively its own systems are functioning.

The agreement establishes specific procedures that could bring the company closer to its goal of knowing more accurately which of its users are under 18. Among other things, it even provides for the use of data that the company previously did not have lawful access to. At the same time, however, it relies on trust in a company which, according to its critics, has often rated its own performance too favorably.

As The New Yorker characteristically notes, Meta has at times given itself an “A,” while the actual results would have corresponded more closely to an “F.”

The Financial Cost and Pressure on TikTok and YouTube

According to an internal company email published in the original complaint in the case, in 2018 a Meta executive estimated the “lifetime value” that a 13-year-old user provided to the company at $270. The core penalties imposed on Meta as part of the case — at least $12 billion — amount to more than $500 for every teenager aged 13 to 17 in the United States.

Although the total amount represents a relatively small share of the company’s revenue — the $12 billion is equivalent to approximately one-fifth of the profits Meta reported for 2025 — the company has essentially agreed to pay more money than it estimates it will earn from its teenage users in the United States.

At the same time, the financial aspect of the agreement is also linked to the stance of other social media companies. Thus, if TikTok and YouTube agree to similar terms and financial penalties, Meta could be required to pay the states that participated in the agreement up to an additional $5 billion.

The existence of this clause leaves room for various interpretations. One is that Meta may have made a calculated settlement: accepting a potential additional payment risk in order to more easily highlight the weaknesses of its competitors.

Meta Calls on TikTok and YouTube to Follow Its Example

Immediately after the agreement was announced, Meta published a letter calling on YouTube and TikTok to adopt the settlement’s most specific and significant provisions.

These include:

the two-hour daily usage limit,
disabling access during nighttime hours,
eliminating notifications during school hours.

Given that Meta had previously strongly opposed the implementation of such policies, the letter is regarded by many as an attempt by the company to turn the agreement into an opportunity for moral superiority over its competitors.

What the Settlement Could Bring to the Rest of the World

The changes agreed upon in the United States are, nevertheless, measures that other countries have also been considering for years. In Europe, Britain, Australia, Brazil, and Singapore, the relevant authorities have developed initiatives aimed at making digital platforms safer for children, often relying on existing laws that require companies to assess risks and protect younger users.

In this particular case, however, the changes did not arise from a federal online-safety law, but from legal proceedings. “This agreement has implications far beyond the United States,” Owen Bennett, an independent technology-policy expert in Britain, told Open Democracy.

“If I were a regulator in Brasília, Brussels, or Canberra, I would see this as a huge opportunity,” he said, noting that the measures authorities have been trying for years to secure from Meta are likely now “a significant part of Mark Zuckerberg’s thinking.” As he explained, because Meta is an American company, pressure from U.S. courts and politicians carries a different weight within the company than rules imposed by foreign regulators. “This agreement will turn the attention of the company’s executives toward child safety in a way that no global regulation or regulatory authority has managed to do to date,” he stressed.

Is Europe Close to Implementing Stricter Measures?

Europe is already putting pressure on Meta over similar issues through the Digital Services Act (DSA), which treats so-called “addictive patterns” in social media as one of the prohibited risks and, consequently, as practices that can lead directly to sanctions.

In April, the European Commission preliminarily found that Meta was violating the DSA because it had failed to adequately identify, assess, and mitigate risks associated with children under 13 accessing Instagram and Facebook.

The Commission said that the company’s existing approach did not effectively prevent minors under 13 from accessing the platforms.

In July, the European Commission went a step further, reaching the preliminary conclusion that Meta was also violating the DSA because of the “addictive design” of Facebook and Instagram. The investigation focused on features such as infinite scrolling, autoplay, push notifications, and personalized recommendation systems — features that were also included in the U.S. states’ allegations. The Commission argued that Meta had not adequately assessed the risks these features posed to users’ physical and mental well-being and that the protective measures it had adopted were not sufficiently effective.

The U.S. agreement is particularly significant because it shows that platforms can change the systems themselves that contribute to creating problems, rather than merely restricting children’s access to them, according to Hannah Storey, head of children’s and young people’s digital rights at Amnesty International. “For years, Amnesty International, civil society organizations, and people affected by social media have warned that major platforms are designed to maximize engagement and profit without adequately protecting children’s rights,” she said.

As she added, the agreement is important not only because of its scale, but because it demonstrates that companies — when held accountable — can change design choices that lead to harm. At the same time, she argues that European authorities already have the tools to move in this direction. “In the EU, a legal proceeding on this scale is not necessary. Existing mechanisms, such as the Digital Services Act, already require platforms to assess and address risks to children.”

A New Point of Reference for European Regulators

The agreement in the United States could serve European authorities as both a point of reference and a challenge. Julia Smackman of the Ada Lovelace Institute told Open Democracy that the U.S. case could strengthen the arguments of European regulators in their dealings with Meta, particularly because of the significant financial consequences that non-compliance could entail.

“Both regulatory interventions and legal proceedings can create incentives for companies to operate more responsibly, provided that the financial consequences are sufficiently large,” she said.

Although the case ended in a settlement rather than a court ruling, and therefore does not establish a legal precedent, the agreement could provide European authorities with a basis on which to support stricter measures. Smackman points out that the European Commission has already proposed measures that could go beyond the provisions of the U.S. agreement, such as disabling features like autoplay and infinite scroll by default.

The Major Challenge: A Safer Design for the Internet

In conclusion, despite the significance of the agreement, much remains to be done to make technology platforms safer, particularly for children.

One of the most important issues remains changing the algorithms that shape content recommendations so that they place greater weight on users’ long-term interests rather than solely on their short-term behavior.

The amount of time a child spends on an app is important, but what content is shown to them is equally important.

For years, the business model of social media has been based on maximizing users’ time and attention. The new agreement — like the parallel legal cases — will not lead to the complete abandonment of this model. However, by increasing the likelihood of new lawsuits and stricter regulations while simultaneously changing the economic expectations associated with companies’ current practices, it creates incentives for platforms to shift their objective: away from extreme usage and keeping users continuously on their apps, toward a more controlled and meaningful form of interaction.