Meta Platforms has agreed to a historic $18 billion settlement with 48 U.S. states and the District of Columbia to implement significant changes to its Instagram and Facebook platforms concerning younger users. The settlement, announced Wednesday, aims to enhance online safety for children and teenagers.
Under the terms of the agreement, Meta will introduce a host of new rules for users under 18. These include strengthening age estimation technology to prevent those under 13 from accessing the platforms and to place teens into age-appropriate accounts. A daily time limit of two hours will be enacted for users under 18, and access to the apps will be blocked between midnight and 6 a.m. Additionally, "like" counts will be hidden for younger users.
While Meta acknowledges these changes represent a significant step forward in child safety, critics and children's advocates express concern that the measures may not fully address the inherent problems with the platforms. Arturo Béjar, a former Meta engineering director, called the settlement a "significant milestone" but cautioned that parents should not consider Instagram suddenly safe for children. He noted that the agreement allows Meta to define harm, drawing an analogy to regulated consumption of harmful substances.
Béjar, who testified during the federal trial that was curtailed by the settlement, emphasized the need for Meta to address the impact of its content recommendation systems on teens, which he believes can be addictive. He pointed out that even when teens stop using the platforms due to time limits, the platform's design can leave them craving more content.
Meta stated that individuals over 18 can voluntarily adopt some of the new controls, such as daily time limit reminders. However, these users will not have the option to opt into the comprehensive settings designed for teens.
To bolster protections for younger users, Meta will enhance its age verification technology, utilizing both internal tools and third-party solutions, subject to regular independent audits. The agreement specifies goals for accuracy in age estimation, particularly regarding false positives where minors are incorrectly identified as adults. If a user is identified as under 13 and removed from the platform, Meta will also assess the ages of their friends.
Béjar praised the age assurance section of the settlement for its inclusion of measurement, independent testing, and specific goals, stating that Meta should be evaluated on effectiveness rather than effort. Meta's current policy requires users to be at least 13 years old to create an account, aligning with federal child privacy law.
Previously, Béjar testified that Meta employed a "don't ask, don't tell" approach regarding users under 13. Despite Meta possessing sophisticated infrastructure for detecting fake accounts, he noted a lack of defined goals or metrics for identifying and verifying the ages of suspected underage users.
Critics have previously faulted Meta for safety features that require users to manually opt-in. Many of the safeguards included in the settlement, such as the default two-hour time limit (which can only be disabled with parental consent) and the nighttime restrictions, will be enabled automatically. However, other features, like turning off video autoplay or viewing posts chronologically, will remain opt-in options.
Josh Golin, executive director of the online safety nonprofit Fairplay, expressed disappointment that the settlement focuses on providing tools to parents rather than outright restricting potentially harmful features. He specifically cited the concern that recommendation algorithms, which can connect children to predators or lead them down harmful content paths, are not turned off by default.
During the trial, Meta's head of Instagram, Adam Mosseri, was questioned about the low adoption rate of the "take a break" feature, which encourages users to pause scrolling. While this feature was made a default for teen accounts launched in 2024, users can dismiss the prompts.
Some research indicates that many of Meta's past safety measures have been ineffective or did not perform as advertised. Yaël Eisenstat, director of policy and impact for the Cybersafety Research Center, noted that it is premature to definitively assess the impact of all desired changes. Her center's recent audit of safety tools from Meta and other companies found that most fell short of expectations.
Despite these criticisms, Eisenstat described the settlement as monumental in demonstrating the feasibility of alternative approaches. She stated that for years, Meta prioritized profit over user safety, disregarding advice from its own employees. The new measures, she added, at least confirm that technically viable options exist to make platforms safer, as advocated by whistleblowers and others.
Eisenstat emphasized that significant work remains, requiring a combination of technical solutions, market pressures, and government regulation. "There's no singular silver bullet for fixing all of this," she said, adding, "I'm just glad that at least we are catching up a little bit now."
The settlement, which also includes U.S. territories, is pending judicial approval.