eSafety Commission urges social media platforms to reveal how they police children

Australia’s online safety regulator has urged the world’s biggest social media platforms to reveal how many children use their platforms and how they enforce age limits. Research conducted by the eSafety Commission shows that nearly two thirds of teens have seen harmful content — including graphic violence and drug use — over the last year, leading Commissioner Julie Inman Grant to call for more debate around the effectiveness of age bans. Changes to the Basic Online Safety Expectations Determination — which grants powers for eSafety to issue transparency notices — led the commission to issue questions to “Google’s YouTube, Meta platforms Facebook and Instagram, TikTok, Snap, Reddit, Discord and Twitch,” the commission announced on Wednesday. The questions come off the back of an investigation by The West Australian, which revealed how a 20-year-old American was utilising Instagram’s algorithm to share violent content to millions of people around the world. Ms Inman Grant said the questions would hopefully produce some “solid data” outlining how many kids were using the eight social media services. “But we also want to assess age assurance readiness and find out how these platforms accurately determine age to prevent children who are under the permitted age from gaining access, and ensure appropriate protections for those who are allowed on the services,” she said. “Most of these platforms have their own age limits in place, commonly 13, but we also want to know how they are detecting and removing under-aged users and how effective this age enforcement is.” Almost a quarter of kids between eight and 10 use social media weekly or more, while nearly half of 11 to 13-year-olds use it the same amount, according to eSafety. Questions on why the eSafety commission did not send questions to X Corp — who recently rejected calls for the removal of graphic content by Ms Inman-Grant — were not answered by eSafety on Wednesday. “eSafety is focusing this first round of information requests on these eight social media services,” a spokesperson said. “eSafety may give further section 20 information requests to more service providers in future.” Speaking with The West Australian, Ms Inman Grant said society had discovered “infinite ways” to misuse technology and that revelations a young man on the other side of the world was peddling graphic content to Aussies, as revealed by The West Australian, was “disappointing”. “Human beings have infinite ways to creatively misuse technology and it’s pretty easy to game the system,” she said. “It’s pretty disappointing to hear about (this on) Instagram because that content would be a violation of their terms of service. “If he’s then putting it behind end-to-end encrypted messaging services like WhatsApp, that would be sort of understandable. “ The eSafety Commission can issue a takedown notice to an online platform to remove harmful online material once it’s reported to the watchdog. According to the commission, 39 removal notices have been issued to companies since powers outlined in the Online Safety Act were introduced. In the last financial year, eSafety received a total of 1340 reports of violent content from members of the public — and 391 of which were classified as containing class 1 material — before they were raised with various platforms for removal. The number of reports relating to violent content received by eSafety has almost tripled since the 2022/23 financial year. Meta claims they “took action” on a whopping 12.1 million pieces of graphic content in the first three months of this year on Instagram alone — which it says was found and actioned by the company itself 99.4 per cent of the time. Meta global head of safety Antigone Davis told an Australian parliamentary inquiry on Wednesday that Meta denied shifting the responsibility for child online safety to the app stores after proposing industry-wide regulation. Ms Davis said children are known to, at times, use “dozens” or even “40 apps” on a device, which she says all pose potential risks. “It is actually trying to solve a problem for parents and make it simpler,” she said. Between April and June this year, Meta claims that between four and five out of every 10,000 content views on Instagram was related to violent content, which compares to one and two out of every 10,000 in 2022. A spokesperson for eSafety confirmed its investigators do not proactively monitor the internet for terrorist or extremely violent material.