Australia’s incoming ban of under-16s on social media can be “privately, effectively and efficiently” enforced, an official report has found.
The Australian government is banning young teenagers from social media from December and commissioned the report to look into how the new rules could be enforced.
It explored a number of methods for verifying users’ ages, including AI facial age estimation, ID checks and parental consent.
The report found that although there was no “one-size-fits-all” approach, there were a “plethora of approaches that fit different use cases in different ways”.
Please use Chrome browser for a more accessible video player
Does new age verification system work?
“This is the most comprehensive and independent evaluation of age assurance we’ve seen,” said Julie Dawson, co-chair of the Age Verification Providers Association (AVPA).
“It shows the technology works, that it can be privacy-preserving and that it is mature enough to give policymakers real confidence.”
Although the findings were celebrated by the age verification industry, the report did find inequalities in how the tech guessed the ages of some groups of people.
Please use Chrome browser for a more accessible video player
Teens targeted with ‘suicide content’
Non-Caucasian, older and female-presenting users faced “reduced accuracy” in certain scenarios, and the fact that indigenous people were underrepresented in training data “remains a challenge”, according to the report.
Teenagers around the age of 16 also faced higher levels of inaccuracy, with 8.2% of 16-year-olds being rejected when they tried to verify they were 16.
Read more from Sky News:
Rudy Giuliani rushed to hospital after car crash
Afghanistan earthquake kills hundreds – all we know so far
In those cases, the user would likely be asked to verify their age another way, either with their ID or with parental consent, according to the report.
“It seems like there are a lot of variations in accuracy,” said Justine Humphry from the University of Sydney, who specialises in online safety.
“That variation is concerning, with the tight schedule for the introduction of a system that will need to be robust and working by the end of this year,” she said.
Please use Chrome browser for a more accessible video player
UK’s online safety rules: One month on
According to the law that passed last November, social media platforms will be liable for fines of up to AUS$50m (£25m) if they don’t prevent children younger than 16 from holding accounts.
The technology explored in the report is the same used by many internet providers in Britain since Ofcom’s new internet safety rules came into force.
The chief executive of Yoti, one of the companies already offering widespread age verification, called the report’s results “very bad news for many opponents who have persistently claimed that even leading safety tech check methods are inaccurate, immature, insecure, privacy chilling & discriminatory.”
