With the Australian Federal Government pushing for a complete ban on social media for children and young teenagers in Australia and a global push for better safeguards on social media, Instagram has introduced a new account type, restricting what kind of content underage users are exposed to.
The new “teen accounts” will kick in automatically for anyone under the age of 18 in Australia, the UK, the US and Canada.
The new rules will limit who can contact these accounts and what content they can see, requiring parental permission to change any built-in protections for those under 16 years of age.
New accounts will have these setting applied immediately, while existing accounts will be transitioned over the next 60 days.
The company claims the timing had nothing to do with the Federal Government’s plans to ban underage users from social media.
"It's going out now because it's ready now, but it's been upwards of a year since we started working on it," head of Instagram Adam Mosseri told the ABC.
"I can't spend a bunch of time trying to design a product with my team that is going to appease a group."
Step in the right direction, but not far enough
Federal minister for communications Michelle Rowland expressed doubt Instagram’s new account settings are unrelated to proposed legislation.
"I note Meta's statement that they've been working on this for some time, and again we welcome that, but I think it also goes to show that decisive actions by governments can have incentives as well," Rowland said.
Yet youth minister Anne Aly said the new restrictions are simply insufficient.
"It certainly does not exclude or abrogate the responsibility of government to keep children safe through having, what we've proposed, age verification and legislation on social media use for children," Aly said.
In a rare moment of bi-partisan agreement, Coalition frontbencher James Paterson was also unimpressed.
"These platforms have known for some time the enormously negative impact they have on the mental health of young people… I still think there is a very strong case for governments to act," he told the ABC.
The core changes for the teen accounts include:
- Restricted messaging settings, allowing messages only from those followed or already connected to.
- Sensitive content control restrictions, limiting content (such as content that shows people fighting or promotes cosmetic procedures) teens see in the more general browsing Explore or Reels sections of the app.
- Teen accounts can also only be tagged or mentioned by people they follow, and Instagram’s hidden words anti-bullying feature will automatically filter out offensive words and phrases from comments and direct message requests.
- Teen accounts will also receive a reminder telling them to close the app after 60 minutes each day.
- Finally, teen accounts will also have an automated sleep mode, which turns notifications off and sends auto-replies to direct messages between 10pm and 7am.
Changes place too much responsibility on parents, say experts
RMIT University professor of information sciences Lisa Given said that while increased protections for youngsters were welcome, Instagram was still shirking responsibility for the content hosted on its site.
“While these controls may be welcomed by some, the onus will continue to be on parents to monitor content and provide support to teens in navigating content safely, rather than Meta implementing new controls to eliminate harmful content at source, to protect all users,” Professor Given said.
“For example, Instagram explains that their Sensitive Content Control applies to some content that goes against their recommendation guidelines, but does not apply to content that goes against their community guidelines.
“Also, content that is limited through Sensitive Content Control is for content from accounts a person doesn’t follow, so content from accounts that are followed remains visible.”
Givens also has concerns about Meta’s enforcement methodology, which proposes the use of AI tools to discern which accounts are breaking the rules.
“Meta also announced it would use artificial intelligence tools to identify children who lie about their age, which may involve age estimation using biometrics (e.g. facial recognition technology) or using behavioural data (e.g. creating an age profile for users based on accounts followed or posts that are liked).
“However, these types of approaches are still being developed and tested, and they have known limitations, including the facial recognition of non-Caucasian users and inaccurate age estimation.”
Meta won’t be launching its AI tools for a while yet at any rate, with plans to roll them out in a trial to US users in 2025.
"There's no perfect solution," head of Instagram Adam Mosseri said. "We are talking about collecting people's biometric face data or government IDs … these are very sensitive pieces of information that I would rather not collect.”