Online Safety Amendment (Strengthening Enforcement for the Social Media Minimum Age) Bill 2026

Ms STEGGALL (Warringah) (18:16): Well, the government is now back in the parliament trying to strengthen enforcement for its social media age ban. The central problems were obvious when the first bill was rushed through this place in 2024: the technology was not settled, the age assurance model was not settled, the privacy implications were not settled and the core legal test of whether the platforms would take reasonable steps was certainly not properly defined.

Those problems remain. I support strong enforcements against big tech. Platforms should not be allowed to hide behind vague community guidelines or the usual Silicon Valley playbook of delay, denial and minimal compliance.

But let's also be really clear about how Online Safety Amendment (Strengthening Enforcement for the Social Media Minimum Age) Bill 2026 came to play and the selective approach the government is taking in what harms they're protecting children from and what ones they are turning a blind eye to. I also want to thank the eSafety Commissioner. I know she works incredibly hard in this space as a local constituent.

I've met with her many times, and I know how hard she worked in this space. This bill gives the eSafety Commissioner bigger sticks and better evidence-gathering powers. It doubles maximum penalties and allows the regulator to compel documents from platforms and third parties.

Those are important tools. The bigger question remains, though, in the billions of dollars of profit these platforms make. Even the doubling of penalties of maximum penalties may be such an insignificant consequence that they are more than likely to want to continue to risk it or use their very deep pockets in circumstances of litigation.

I have strong reservations around how much more of a bandaid this is to a bigger problem. What we know is that this legislation and these amendments do not fix the central weakness in the scheme: what 'reasonable steps' for age verification actually means in practice. The government calls this a world-leading ban.

Well, a world-leading child safety framework would start with safety by design, safer algorithms, safer default settings, stronger privacy protections, transparency over recommender systems and a real statutory duty of care on platforms. The Prime Minister and the Minister for Communications are already spruiking this social media age ban as a success on the international stage despite the evidence showing that it has had limited effect in terms of actually keeping young people off and protecting them from harm.

We have other countries now following Australia's lead whilst we're still working on the legislation that would actually make social media safer: a European Digital Services Act-style model focused on platform accountability and safety by design. The government continues to promote this bandaid fix internationally when the evidence so clearly points to a systems focused European Digital Services Act-style model as being far, far superior.

This will be what I'll be looking for when the government introduces its long-awaited promised digital duty of care framework later this year—a best practice framework not watered down as we have seen and has occurred in other jurisdictions. A ban can only ever be one part of the answer, or you just push the problem underground. In Warringah parents, teachers and young people have been very clear with me in their feedback.

They're worried about what children are exposed to online. They're worried about the addictive design. They're worried about the algorithms pushing harmful content.

They're worried about cyberbullying, eating disorder content, misogynistic material, scams, AI deepfakes and the relentless pressure children feel to post their lives online. But young people have also told us something else: that social media is not only a space of harm. It is also where they communicate, organise, seek, help, build communities and have a voice.

When I spoke to the original legislation in 2024, I referred to Lucy Flynn, a 14-year-old from Warringah who had recovered from an eating disorder and used social media to petition for more public hospital beds for eating disorder treatment. Lucy harnessed the power of social media to tell her story and amplified the issues in the eating disorder funding. Online platforms can provide a platform to reach people outside their geographical communities, provide an avenue for youth leadership on issues that matter to them and provide the ability for young people to feel that they have control over their future.

So I disagree with the government. The answer is not a blanket ban on social media. It is to make social media socially safe by design, fit for the digital age and for young people like Lucy to pursue their passions and find their tribes.

A cynical hypocrisy exists when a government lets gambling ads harm the wellbeing of Australian children—children as young as six can identify gambling company brands—while posturing here in this place to be cracking down on social media tech giants with an ineffective and, I would say, disingenuous age ban for social media platforms. The social media minimum age scheme requires age restricted platforms to take reasonable steps—this is important for people to understand what this legislation actually does—to prevent children under 16 from creating or keeping accounts.

The platforms covered include major services such as Facebook, Instagram, Snapchat, TikTok, YouTube, X, Threads, Reddit, Twitch and others subject to the statutory criteria. Keep in mind, there are a lot of concerns around gaming platforms that also create cause for harm. What it's really important to understand is there is no narrowing of that definition around reasonable steps, and I'll come back to that in a moment.

The government says this new bill will strengthen eSafety's information-gathering powers and double maximum penalties for systemic breaches from $49.5 million to $99 million. Again, I say, for companies that are making billions of dollars of profit, there is the question of whether that is even meaningful in terms of a stick or a penalty. This amendment will allow eSafety to compel evidence from platforms and third parties, including age assurance and app store providers.

That is what this bill provides for. These are simple changes and fine in their own right if one believes that the system actually works. But the context of this debate does go back to the original bill in 2024.

Then, crossbenchers and experts noted that it did not specify how the ban would be implemented and that 'reasonable steps' was not defined in the legislation. It continues to not be defined. I've asked that of the minister, and the explanation was, 'We don't want to be held to it now and to allow for that to continue changing.' But what it does do is create a huge legal grey area.

The Senate inquiry into the original bill recommended that the government legislate a digital duty of care, meaning a positive legal obligation on platforms to proactively identify, prevent and reduce online harms. Instead, the government rushed through a blunt age based scheme before the age assurance trial had reported, before the practical enforcement model was clear and before the parliament had properly tested for unintended consequences.

I said at the time that the bill was misguided, that it was window dressing and that it risked driving harms underground while failing to hold platforms properly accountable. I also said the government was putting the cart before the horse by legislating first and working out the detail later. But, of course, here we are, fine-tuning some detail.

The recent evidence bears all of those concerns out two years on. A University of Newcastle study found little immediate evidence that the restrictions had reduced access, with more than 85 per cent of surveyed under-16s still using restricted platforms after the policy began. Young people reported bypassing age checks through fake accounts, friends or family, incognito browsing and other workarounds.

Experts have made the same point: a simple ban is unlikely to be enough if the underlying causes of harm, including platform design, algorithms and age-inappropriate features, are left untouched. There are also now live constitutional challenges in the High Court—one by Reddit, another by teenagers—arguing that the scheme burdens the implied freedom of political communication and that less restrictive alternatives were available, including safety-by-design tools.

That doesn't mean the challenges will succeed, but it does show the legal risks were foreseeable in the event they do. The constitutional challenge to the social media minimum age scheme cannot be viewed in isolation from the enforcement issues this amendment bill seeks to address. They're linked by the same unresolved problem: the act still does not clearly define what reasonable steps are actually required of the platforms.

If the Prime Minister thought a royal commission into domestic violence would be a lawyer's picnic, let me tell you that this bill is a lawyer's picnic, especially when you're talking about social media platforms with extremely deep pockets and an extremely well-developed desire to contest these kinds of restrictions. Without a clearer definition of 'reasonable steps', the government is simply inviting exactly the kind of litigation it says it wants to avoid.

Platforms will argue over what was reasonable, lawyers will test the boundaries of the act, and the courts will be left to do what the parliament should have done in the first place. The government should not mistake stronger enforcement powers for a stronger policy. I support giving the eSafety Commissioner the tools it needs to investigate powerful platforms.

I support higher penalties where companies fail to comply. I support holding big tech accountable. But I do not accept the fiction that the original scheme was well designed and now merely needs some heavier penalties.

The obstacles were obvious then: vague obligations, unsettled technology, privacy risks and the absence of a proper digital duty of care. Parents deserve better than being told that the problem has been solved when they can see every day that it has not. Young people deserve better than a policy that treats them as the problem rather than focusing on the platforms that profit from harmful design.

The proposed digital duty of care must be the centrepiece of the next stage of reform: a best-practice framework that requires platforms to design for safety, gives regulators visibility over the algorithms shaping children's lives and backs prevention with proper enforcement, education and support. If not, it is just more window-dressing. (Quorum formed)