Teens, it’s said, live on their phones, and one of the places they spend the most time is Instagram. For years, the perception has been that they’re completely unsupervised there, to their detriment. That could be changing: Instagram’s owner Meta announced today that teens using the app will face a series of new restrictions, as well as increased parental controls. Under the new policy, accounts created or owned by anyone under 18 will have limited functionality by default — an effort, the company says, to give parents “peace of mind that their teens are safe with the right protections in place.”

These changes, many would argue, are long overdue. For years, concerns have been raised about the effects that rampant, unsupervised social media use can have on young people—that these platforms can contribute to depression, anxiety, serious body image issues, and even suicide risk. Meta has come under scrutiny, particularly since former Facebook data scientist Frances Haugen leaked a trove of internal documents in 2021, some of them related to teens’ experiences on Instagram and Facebook. Subsequently, Meta and other social media companies were hit with a wave of lawsuits alleging harm the platforms caused to adolescents, with politicians on both the right and the left suggesting that it might be a good idea to require parental consent for children’s use of algorithmic feeds, or to ban younger teens from using social media altogether. “Facebook is not interested in making meaningful changes to improve the safety of children on its platforms,” Marsha Blackburn, a Republican senator from Tennessee, said during one of Haugen’s convention appearances. “At least nothing that would result in fewer people watching the messages or a decrease in their advertising revenue.”

Now, Meta is trying to prove otherwise. The new protections will almost certainly make Instagram less attractive to minors. The new “Teen” accounts are private by default, meaning their posts can’t be viewed by anyone who isn’t an approved follower, and they can only receive messages from accounts they follow “or are already connected to.” Teens will also be prompted to close the app after 60 minutes of use, and their accounts will automatically be put into “sleep mode,” which mutes notifications, from 10 p.m. to 7 a.m. (The app will still be usable during these hours, though, and teens can choose to check their direct messages at any time.) Additionally, these accounts will be subject to Instagram’s “most restrictive” content filter by default, and anyone under 16 won’t be able to change the setting without their parents’ permission. (Meta already enables this filter for younger users, but until now those users have been able to change it themselves.) In a press release, Meta notes that the filter is intended to limit teens’ exposure to content that shows, for example, “people fighting” or promotes “cosmetic procedures.”

There is some flexibility built into the new system. While Instagram is enrolling all teen users under the age of 18 in its new program, users 16 and older will be able to change their default settings, such as turning off sleep mode. (Clearly, teens can simply lie about their age when creating an account; more on that later.) If younger teens want to make changes, they’ll need to add a parent or guardian through the app’s settings and get their permission to make any changes. Additionally, there’s a comprehensive parental controls tool, which both the parent and teen must opt ​​into, that lets parents and guardians see who their teen has messaged in the past seven days (though they won’t be able to read the content of those messages), set daily time limits for the app (which can be enforced with a pop-up reminder or a hard shutdown), and block Instagram from being active for preset periods of time during the day (like school hours).

“On the surface, it’s what a lot of people have been advocating for a long time,” Candice L. Odgers, associate dean for research and professor of psychological science and computer science at UC Irvine, told me. (Earlier this year, Odgers wrote a paper for The Atlantic Ocean (I spoke to her about her research on children’s use of digital technology, in which she argued that extreme rhetoric about the supposedly harmful effects of social media can itself be harmful.) I gave her only a brief summary of Instagram’s update, the details of which weren’t public when we spoke; she noted that the default settings may be the most significant development here. “We’ve said for a long time that this is too much of a burden that falls on parents, on teens, on caregivers,” she told me. “The closer we can get to safety by design … the better it will be for everyone.”

Read: Smartphone panic isn’t helping teens

Jonathan Haidt, a social psychologist at NYU’s Stern School of Business and one of the most prominent and influential voices on teen social media use, has argued that these apps are downright “dangerous” for young people. His latest book, titled The Anxious Generation: How the Great Reprogramming of Childhood Is Creating an Epidemic of Mental Illnessargues this extensively and is a bestseller. (Haidt is also a regular contributor to The Atlantic Ocean.) But even he saw today’s changes as a “huge step in the right direction” and told me he’s “very encouraged” by them. “Meta is the big fish,” he said. “I think Meta being the first to do it is a really good sign, and it will probably encourage other platforms to treat teens differently.” (He qualified his optimism by saying that young people will probably still be on their phones far too much. But “at least it would take away some of the worst things, like being approached by strange men.”)

Liza Crenshaw, a spokesperson for Meta’s youth and wellness team, told me the company was focused on responding to parental feedback when it created the features. “I don’t really know if we’re trying to solve anything other than what we’re hearing from parents that they want,” she told me. She said the company has interviewed parents who have asked for more ways to be involved in their teens’ use of the app, and they wanted the safety features turned on by default.

Odgers praised that approach in theory, noting that as a parent, it’s better to have the most restrictive settings yourself and then talk about lifting some of them over time and through negotiation. But, she added, “the devil is in the detail.” The most obvious problem with Instagram’s new approach is age verification. The company has already experimented with using facial analysis software to guess users’ ages and apply some restrictions to accounts with younger users. The next step is to use it to categorize accounts as “teen” in situations where a younger teen might be trying to bypass restrictions. If the system thinks a user is lying about their age, it could ask them to verify their age with an ID or through a “facial estimation tool,” which involves an AI analysis of a person’s “biometric selfie.”

Jonathan Haidt: Get Phones Out of Schools Now

There are other problems, too. To prevent, say, a 20-year-old from signing up as a parent-guardian who can give permission to a dozen of his 16-year-old sibling’s friends to turn off restrictive settings, Meta says it will cap the number of accounts any one person can manage. Crenshaw declined to be specific about what that limit will be, and said Meta won’t share all the ways it plans to combat circumvention of the parental controls tool. She also acknowledged that the company will have to be careful not to penalize teens who don’t have a traditional parent-guardian in their lives, but said she couldn’t provide details about how Instagram will make those distinctions.

Still, many concerned parents see these characteristics as cause for celebration. Meta certainly does. This morning, the company is holding a three-hour announcement event at its New York office, hosted by actress Jessica Alba, and Thursday night there’s a party at Public Records, a popular Brooklyn music venue. The festivities could be in honor of prevailing common sense: Scientists largely agree that there’s enough evidence to raise concerns about the link between social media use and depression and anxiety, particularly among younger teenage girls. They disagree about what to do about it, but today’s measures represent a good-faith attempt to do something.