Australia finds serious gaps in Big Tech’s response to child sexual abuse online

Australia’s online safety regulator has told the largest technology companies in the world that the tools to stop sexual extortion already exist, and that they are not using them. In a transparency report published on Tuesday, eSafety said companies including Apple, Meta, and Google have “significant gaps” in how they detect and prevent child sexual exploitation and abuse.

The specific failure named is language analysis. Sexual extortion offenders work from recognisable scripts, the same coercive phrases repeated across thousands of approaches, and the report says platforms are not deploying the technology that would spot them.

“In several cases, we have provided these platforms with evidence of how their services are being colonised by criminals to devastating impact, with clear guidance on how to stem the abuse,” said eSafety Commissioner Julie Inman Grant.

“Even when we’ve laid this out, we haven’t seen adequate responses, despite the technology being readily available.”

Google, Meta, Snap, Microsoft, and Apple did not immediately respond to requests for comment.

The report is the third in a planned series of four, and this one is built around sexual extortion, a form of blackmail in which offenders threaten to publish intimate images unless a victim pays or complies. The regulator received more than 2,000 complaints about it between July and December 2025.

The demographic most affected is not the one most people picture. It is young men aged 18 to 24, but teenagers are not spared. An eSafety study last year found more than one in 10 people aged 16 to 18 had been victims of sexual extortion, and more than half of those had been targeted before they turned 16.

The reporting tools themselves are part of the problem. Gaps persist across services including WhatsApp, iMessage, Discord, and Google Messages, the report found, with some lacking a clear way to report sexual extortion or child abuse at all, and others offering no dedicated category for it.

Livestreamed abuse is the other unclosed door. Technology to detect it exists, the report says, but it is not being consistently deployed across the services where it would matter.

Not everything in the document is a rebuke. Google and Snap have taken steps to proactively detect known child sexual abuse material, Discord has begun blocking links to it, Meta has introduced new tools to detect grooming, and Microsoft is now detecting live abuse in video calls.

The powers behind the report date to 2024, when eSafety directed eight technology companies to report every six months on their compliance with Australia’s Basic Online Safety Expectations. The first report set a baseline.

The second found companies were failing to proactively detect abuse material. The third says they have been told exactly what to do and have not done enough of it.

The regulator has been circling adjacent ground for months. In April it asked online gaming platforms to explain how they protect children from grooming by predators, a question the industry has been answering unevenly, with Roblox splitting its users into age-gated tiers under pressure from child safety lawsuits.

It also arrives while Canberra is trying to give eSafety sharper teeth. Legislation introduced in June would let the regulator pursue platforms in court over non-compliance with the under-16 social media ban, part of a package that would roughly double the maximum penalty for a systemic breach.

That ban, the first of its kind, is not going smoothly. eSafety has already accused Meta, TikTok, and YouTube of failing to take the reasonable steps the law requires, and Britain and several European countries are watching the outcome while their own child-safety rules stall in negotiation.

The fourth and final report in the series is still to come. On the evidence of the third, the companies have had the guidance, the data, and the technology for some time.