Australia's teen social media ban has officially begun. Here's what's changing

After more than a year of planning and global scrutiny, the teen social media ban is finally a reality for Australians.

Here's everything you need to know about the policy, from the ABC's national technology reporter Ange Lavoipierre.

And if you've got a question or story to share, you can contact us using this form. 

What is the teen social media ban?

Believe it or not, this is not actually a silly question, because the new rules aren't what most people think.

In short, the "ban" is not exactly a ban.

An amendment to the Online Safety Act in November 2024 will force some (but not all!) social media companies to prevent (or try to prevent!) under-16s from holding an account on their platforms (but not from using them!).

The bits in brackets often get missed in the public debate, but the reality is that this policy won't actually block all under-16s from all social media, even if it does work as designed.

And to be clear, no-one, including the government, is expecting a flawless start.

That said, I'm still going to call it a "ban" because, for better or worse, that's the name that's stuck.

So from the top, here are the most important things to know for now: what the new rules will and won't do, and what they mean for you.

Which platforms are included in the ban?

At the time of writing, the law applies to only 10 of the bajillion or so websites in the world you might, if you were feeling generous (or cruel, perhaps) describe as "social media".

So far, the platforms in question are:

• Facebook
• Instagram
• TikTok
• Snapchat
• X (formerly Twitter)
• YouTube
• Reddit
• Twitch
• Kick
• Threads

Already, there are signs plenty of under-16s are planning to move to other platforms that aren't covered by the new rules, such as Yope, Lemon8 and RedNote.

But there's no guarantee those sites will be left out of the ban forever, with the government warning the list could grow, setting the stage for the world's most expensive and annoying game of whack-a-mole.

Proving the point, Reddit, Kick and Twitch were only added in November.

No matter what happens to the list, it seems as if under-16s will still have plenty of places to gather online, even if they're not the places they're used to.

What will the platforms have to do?

Politicians love to say that platforms must comply with the laws "or face $50 million fines" but compliance doesn't necessarily mean catching every single user under-16.

The law only says that platforms must take "reasonable steps" to prevent under-16s from holding an account.

If people bypass the ban, it's not a fine-able offence.

As for what constitutes "reasonable steps", it boils down to a platform checking every user's age, or being confident they don't need to, which you could argue is the same thing.

There's more detail about what is meant by age checking below, but the main thing to know is that social media companies don't actually need to be certain of your date of birth.

The rules are clear there's no need to "verify" age. The platform only has to make a well-informed guess.

The other reason you could argue this isn't truly a ban is the platforms don't have to keep young people from looking at the platforms altogether — it's all about who's allowed to hold an account. There's nothing in the law to prevent under-16s from accessing those websites in what's called a "logged-out" state.

Most of the platforms, including YouTube, TikTok, Instagram, Facebook and Reddit, will let you see a fair bit of content without an account, although they'd all say the most harmful stuff will remain hidden.

Perhaps the key difference is that logged-out users aren't fed content that's been optimised to keep them glued to their screens for hours on end, in many cases by learning their preferences and ruthlessly exploiting their insecurities.

How are the platforms going to check your age?

Platforms have been given a menu of options for checking your age, and they'll all offer more than one.

The most accurate option on the table is age verification, which almost always involves handing over government-issued ID, such as a drivers licence.

The government has been at pains to make clear that the law forbids ID from being the only age-checking method on offer, but the law is less clear as to whether ID is allowed to be the last resort, when other age-checking methods fail.

Another option, and one which is shaping up to play a big role, is facial-age estimation.

In this scenario, you agree to let AI software guess your age based on your physical characteristics, usually by scanning your face.

Having seen it in action, it's not always accurate — it can be wrong in either direction. Even industry proponents say it shouldn't be relied on as a sole solution, especially for people closer to the cut-off age.

The third main method is age inference, where a platform guesses your age based on what it already knows about you — things like how long your account has existed, your posts, your interests and your personal networks.

That said, there are privacy rules against platforms being too creepy in how they go about this.

Companies aren't allowed to bundle too much information together and, no matter which method they use, they have to delete the data after they've used it.

And hey, if any of this feels confusing, you're not hallucinating!

Every platform is using a different combination of methods and service providers to check our ages.

For example, YouTube, Meta and Snapchat have all announced their plans in recent weeks.

What will change for me?

For under-16s, the impact is too obvious to mention.

For many Australian adults, however, it's possible very little will change.

What's becoming clear is that many social media platforms are quite happy to use the data they already have about existing users to rule them in, when they're well and truly over 16, which means those people might never be asked for more data.

On the other hand, anyone starting a new account, and some existing users over 16, especially those closer to the cut-off age, will be asked to demonstrate their age another way — probably with ID or a face scan.

For anyone who either can't or won't do that, things might get bumpy.

Platforms are obliged to provide avenues for appeal in case they get it wrong, but we're still waiting to see how smoothly those systems will work, and whether they might ultimately depend on people providing the exact same data they couldn't or wouldn't provide in the first place.