Hello! brave enough is back after a brief period of Berlin-winter-induced writer’s block. Thanks for sticking around. I’m excited to share some new essays with you.
Last week proved historic for Meta, for all the wrong reasons. First, a court in New Mexico found the tech giant liable for endangering children on their platforms. One day later, a young woman in Los Angeles successfully sued both Meta and YouTube over her childhood addiction to social media. While Meta will undoubtedly pursue appeals, these decisions signal a definitive shift in the legal landscape regarding child safety online. I suspect that these landmark court cases will fuel stricter legislation on social media age restrictions for children around the world.
Traction on this topic has been building for years now, with children’s safety campaigners and books like The Anxious Generation leading the charge. And while these shifts might have slipped past you given the relentlessness of the news cycle lately, global governments have already started moving from debate into active legislation.
Here’s a brief summary of what’s happened in several countries over the last three months:
Australia enforced a total ban on social media for under 16s in December 2025
France passed a ban on social media use for under 15s in January 2026
Spain announced plans for a ban on under 16s using social media in February 2026
Malaysia will enforce a ban for social media for under 16s in July 2026, along with mandatory eKYC (electronic ID) verification
The UK is currently implementing the Online Safety Act, which will require “highly effective” age checks for social media usage
With stricter legislation in place, tech companies would eventually be forced to implement changes to their architecture, integrating “safety-by-design” principles. Some common safety-by-design proposals include:
Private profiles by default
Disabling infinite scroll to prevent doomscrolling
Disabling likes to protect against addictive dopamine hits
Zero personal data collection by the platform
Curated versions of platforms that hide adult content (e.g. YouTube Kids)
Heavily restricted direct messaging from strangers
Daily time limits that can only be adjusted by a parent
Allowing parents to opt out of recommendation algorithms entirely, showing only posts from friends
This would likely be combined with some form of age assurance technology to verify a user’s age with hard proof.
All of this is drastically different from the practices of the last decade and a half, where age restrictions were pretty much limited to the “check this box if you’re over 16 – we trust you!” method.
Unsurprisingly, the status quo has failed miserably to keep the kids safe. We’ve been cracking jokes about iPad kids for years now, so these changes ought to be something to celebrate… right?
While the intent may be noble, the recent scramble for stricter age verification has led to a lot of pushback from concerned campaigners and journalists who warn that these reforms are a Trojan horse – one that opens the door for mass surveillance.
At the risk of sounding like a tinfoil-hat-wearing conspiracy theorist, they’re not wrong. We need only look at OpenAI’s recent decision to partner with the Pentagon to understand how quickly technology designed with “human interest principles” can be weaponised. Altman’s move proved to be so controversial that the QuitGPT movement was launched.
My question is: why aren’t we as skeptical about what’s happening with social media?
Earlier this month, Meta announced they will remove end-to-end encryption on Instagram messages from May 2026. This means they will be able to see the contents of all messages between users, a decision seemingly triggered following years of pressure from child safety groups and law enforcement, including the FBI and Interpol. The problem is that big tech directly benefits from these decisions via massive amounts of data to harvest on users. Interestingly, Meta’s spokesperson cited low adoption of end-to-end encrypted messaging as the reason for the change. While I don’t doubt that this is technically true, it is somewhat alarming that the privacy-first message that Meta was hocking over the last few years now seems to have gone out of style. Even if they were to offer this update for underage accounts, there is no excuse for rolling back encryption for adults.
Giving the government the ability to link your identity with movements you support, causes you’ve donated to, and other highly personal information could have disastrous consequences – depending on who is in power. This is particularly pertinent for activist groups, grassroots movements, and discriminated communities, for whom social media is vital for spreading messages that aren’t always seen on mainstream channels. I can’t help but think: with their identity linked to their account, would an activist in Taliban-controlled Afghanistan be brave enough to post photos without a burka in protest? Would an American teenager risk searching for information on reproductive healthcare? Would a lonely member of the LGBTQ+ community feel safe enough to connect with peers they might desperately need?
Two things can be true at once. The best and worst thing about the internet is that it can be anonymous. Perhaps it’s very European of me, but being able to use the internet privately seems crucial to avoid a complete surveillance state – even if the same logic allows others to conduct dark business.
That said, the concerns raised by child safety groups are also valid. In my view, the only way forward is to try and navigate the middle ground, finding a solution that protects children but also puts guardrails in place to protect personal data more generally. Neither the unregulated digital Wild West of the past nor the Orwellian internet of the future is ideal. Getting it right actually has everything to do with protecting children – I fear that the stricter identity verification becomes, the more it could push young people into darker, less-regulated corners of the internet.
When I reflected on a hypothetical middle ground, my thoughts immediately jumped to decentralised age verification. This would require a trusted third party to provide users with an encrypted digital ID, which they can then share with the platform to prove their age without revealing any personal information. Unsurprisingly, the EU is leading the charge here with its Blueprint for Age Verification, an initiative expected to be fully interoperable with the EU Digital Identity Wallet by late 2026.
Another option is device-level verification, which would push the burden from apps to the hardware itself, where your phone’s operating system stores an encrypted age signal. There’s also lots of innovative companies out there, such as Yoti, who are developing privacy-preserving age assurance technology. Yoti’s tech uses AI to scan a face and estimate age within a range, without ever knowing the person’s name or storing their image.
The wonderful thing about technology is that there is always a way. It just depends on whether the public steps up to demand better.
At the start of this essay, I referenced Jonathan Haidt’s The Anxious Generation, a bestseller which has sparked a global movement to end so-called “phone-based childhoods”.
Haidt’s book hasn’t escaped scrutiny. In this Nature article, developmental psychologist Candice Odgers argued that Haidt’s claims are not supported by rigorous scientific evidence. Despite this critique, I believe that the books’ central thesis still holds true: in the 21st century, we under-protected children online while overprotecting them in the real world. And while I agree with many of Odgers’ sentiments, I also don’t require a peer-reviewed study to believe that Facebook caused harm by selling the precise moment a 13-year-old girl deletes a selfie to companies advertising flat tummy tea.
Whether we like it or not, our online lives weave into our lived experience. If we can accept that the internet is a place – albeit a digital one – where we spend time, then it’s crucial to develop a sense of judgement and common sense about it. It’s also crucial that we are afforded a degree of privacy.
Both UNICEF and the LSE argue that the internet is a skill that must be taught, not a danger to be avoided. Some schools in Australia are trialling “digital licenses,” where students must pass modules on cyberbullying and privacy before being granted access to school devices. Such initiatives, combined with age guidelines for owning a smartphone or having access to social media, seem like a reasonable way forward. In my view, we should treat social media access like drinking alcohol or driving a car: not necessarily an evil thing, but something serious that requires a certain level of maturity.
I really feel for anyone who is currently a parent trying to manage this in real time. It’s easy to think about the topic in an academic, cerebral way – yet I’m sure it’s another thing entirely when your child is begging you for a smartphone “because everybody else at school has one”.
We need to analyse the decisions governments make in the coming years very closely. Most modern technology develops in a way that simply happens to us – we tend to accept the deal before we’ve understood the costs. As we move toward a future of stricter verification and safety-by-design, we must ensure that in our rush to protect children’s innocence, we don’t sacrifice the privacy that protects their future autonomy.
Thanks for reading – I’d love to hear your thoughts in the comments!
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.