4 hours ago
5
When the age assurance technology trial released its final report before Australia’s under-16s social media ban came into effect last year, its first finding was: age assurance can be done privately, efficiently and effectively.
Four months since the ban came into effect, we can say that was – to paraphrase Yes Minister – a courageous statement.
The country’s eSafety commissioner revealed on Tuesday that more than two-thirds of teens were still on the platforms included in the ban, children were easily bypassing facial age estimation technology if aged within two years of 16, and half of the platforms initially included in the ban were being assessed for non-compliance.
eSafety also found 66% of parents whose children remained on social media said that the platforms had not asked their child to go through age verification, while others reported that if the age on an account was said to be 14 or 15, the platforms asked users to go through facial recognition and adjust their age rather than deactivating the account.
Many of the complaints to eSafety were from parents who had reported their teen’s account to the platform, only for the platform to take no action. The report noted Guardian Australia’s reporting had confirmed this.
Many of those who had been pushing for the ban had simply accepted the technology trial report in terms of how easy it would be to achieve, but those who actually delved into the data noted at the time that facial age estimation would be less accurate for those aged 14 and 15 years old – exactly the age group the platforms were supposed to be blocking.
We can now see, by eSafety’s own reporting, “facial age estimation is known to have higher error rates for children near the age threshold of 16 years”, as many had warned. But age assurance companies had talked down that possibility when asked last year.
The ban has also had unintended side effects for the government. On the day new research found vaping could lead to cancer, Guardian Australia exclusively obtained documents from the health department that showed some of the social media portion of its anti-vaping ad campaign targeting teens had to be diverted to gaming platforms and audio platforms such as Spotify to reach 14 and 15-year-olds.
Just 10% of the ad spend was diverted to gaming to reach this demographic – despite many of those still using social media.
The Albanese government has been keen to sell its early success on the social media ban to the world – with more than 5m accounts now deactivated – but these early problems should give other countries pause for thought before following.
The communications minister, Anika Wells, said on Tuesday she expected eSafety to “throw the book at” non-compliant tech platforms. That would mean launching court action in coming months to seek A$49.5m in fines for a breach. But, notably, Wells did not answer a question put to her regarding when that would be.
That could be because high court challenges over the validity of the social media ban – one from Reddit and another from a digital rights group – have yet to be heard. The federal government has now filed its defence in the latter case, but it’s unlikely the matter will be heard until later this year.
If, ultimately, fines are issued to bring the tech companies to heel, there is the possibility we could see a big overcorrection, with platforms switching to more intrusive methods of validation – not just of teens but all new accounts – such as using ID checks.
The platforms had already designed specific safety features and parental controls designed to keep teens safe online. Since the ban, the teens who have managed to bypass age checks are no longer given the bare minimum of safety features the platforms have included for teens, given their account now appears to be held by someone older.
In the meantime, other countries should consider waiting for more data on the effectiveness of the ban from the study commissioned by eSafety surveying 4,000 teens and parents in February. That too, is not without issue. Crikey reported last week that of the thousands who had signed up to participate, just 273 had opted into using apps that would track their app use.
Other internal eSafety documents seen by Guardian Australia noted this form of tracking was more accurate than self-report surveys.
On Tuesday, Wells said: “Australia has started a global movement. More than a dozen countries now are following our lead.”
Perhaps other countries should consider waiting until Australia sees if the ban is actually achievable.
