The Circuit Breaker

With 4.7 million accounts purged in a nation of only 2 million children, Australia’s digital safety experiment reveals a chaotic reality.

[Speaker 1]: On January 15th, the Australian government released a number that was supposed to end the argument. They announced that in just over a month-since the social media ban came into effect-4.7 million accounts had been deactivated or restricted. [Speaker 2]: On paper, that looks like a victory. That is a massive purge of under-16s from the digital ecosystem. [Speaker 1]: Right. If you’re a parent who was promised a "circuit breaker" for your kid’s social media addiction, you look at that 4.7 million figure and you think, "Okay, the system is working." [Speaker 2]: But then you look at the census data. And you realize that in the entire country of Australia, there are only about 2 million people aged 10 to 16. [Speaker 1]: So, we have more than double the number of bans than we have actual children. [Speaker 2]: Which begs the question: Who, or what, did Australia actually ban? And did this "world-first" experiment actually make the internet safer, or did it just break the parts that were working? [Speaker 1]: It’s Wednesday, January 28, 2026, and you’re listening to The Angle. [Speaker 2]: So to understand how we got to these numbers-and the mess we’re seeing right now-we have to go back to late 2024. That’s when this really shifted from a debate into law. [Speaker 1]: The rhetoric back then was incredibly specific. Prime Minister Albanese and Minister Wells weren't talking about censorship; they were talking about exhaustion. The pitch to parents was essentially, "We know you can't police this alone. Let the government be the bad guy." They called it a "circuit breaker" for mental health. [Speaker 2]: And the mechanism they built to achieve that was the *Online Safety Amendment Act*. And the most critical thing to understand about this law is where the liability sits. There are zero fines for kids who sneak on. Zero fines for parents who help them. [Speaker 1]: The gun is pointed entirely at the platforms. [Speaker 2]: Exactly. The law says that if Meta, or TikTok, or Google fails to take "reasonable steps" to prevent under-16s from having accounts, they face fines of up to 49.5 million Australian dollars per systemic failure. [Speaker 1]: That number-49.5 million-is the engine driving everything we’re seeing today. Because when the ban actually kicked in on December 10, the tech companies didn't all react the same way. They looked at that fine, looked at their own architecture, and scrambled in completely different directions. [Speaker 2]: And that’s where the chaos started. Because the law didn’t mandate a specific technology. It didn’t say "everyone must upload a driver’s license." It just said "age assurance." [Speaker 1]: Which sounds like a technicality, but in practice, it’s a massive distinction. "Verification" is proving who you are. "Assurance" is just proving you aren't a child. [Speaker 2]: Right. And since most platforms don't want to collect millions of passports-because that’s a privacy nightmare and a security risk-they leaned heavily on what’s called "Age Inference." Basically, using AI to look at your behavior, your friend networks, even the way you type, to guess how old you are. [Speaker 1]: So fast forward to today. We are seven weeks past Implementation Day. The government is claiming victory with that 4.7 million number. But when you peel back the layers of how the different companies actually complied, the "safety net" looks really inconsistent. [Speaker 2]: Let's look at Meta first-Instagram and Facebook. They took the most aggressive route. They used what they call an "Adult Classifier." It’s an algorithm that runs in the background,…