Former Australian of the Year Grace Tame has emphasized the urgent national need to stop AI tools from being used to create child sexual abuse material, urging the government to make possession of child exploitation software a criminal offence.
Ms. Tame and other child safety advocates are meeting at Parliament House today, just before the new parliamentary term, to discuss the growing use of AI in child sexual exploitation. They will also explore how AI can be used to detect grooming and abusive content. This comes amid increased public scrutiny following another disturbing alleged abuse case at a childcare centre in Melbourne.
Ms. Tame, known for her advocacy on child abuse survivors’ rights, criticized both past and current governments for slow action on online child safety.
“I don’t think previous governments — and unfortunately, the current government — have acted fast enough to protect children online,” she said. “We understand the risks. We’re waiting for urgent and effective government intervention.”
The International Centre for Missing and Exploited Children (ICMEC), which is leading the parliamentary roundtable, has called on the government to make it illegal to possess or distribute AI software specifically built to generate child sexual abuse material (CSAM). Similar legislation has already been introduced in the United Kingdom, which Australia is reportedly monitoring.
According to intelligence firm Graphika, explicit AI tools that were once limited to fringe online communities have become widespread and commercialized. A 2023 report revealed that 34 such tools had received over 24 million visits, with growing availability across platforms like Reddit, X (formerly Twitter), and Telegram. Law enforcement agencies say these tools are draining resources that should be directed toward cases involving actual victims.
Australia’s 10-year national child protection strategy, drafted in 2021, does not currently address the threats posed by AI. ICMEC Australia CEO Colm Gannon said this gap needs urgent attention:
“This software offers no benefit to society. It should be regulated, outlawed, and criminalized. The urgency is real.”
Ms. Tame highlighted that offenders can easily access and download AI tools to create abusive content offline, making detection extremely difficult.
“It’s like the wild west — and it doesn’t take much technical skill,” she warned.
An independent review of the Online Safety Act, submitted to the government in October 2023, also recommended banning “nudify” AI applications that create non-consensual explicit images. While the government has pledged to enforce a “duty of care” on platforms to protect children, this has yet to be legislated, and most of the report’s 66 recommendations remain unaddressed.
Attorney-General Michelle Rowland described the use of AI to generate child abuse material as “sickening” and stated she is committed to enhancing the government’s regulatory response, especially in high-risk AI applications.
Advocates also want the government to ease restrictions that limit law enforcement’s ability to use AI to investigate child abuse. Since 2021, authorities have limited use of facial recognition software like Clearview AI after the Privacy Commissioner ruled it had violated Australians’ privacy by collecting biometric data without consent.
However, Gannon noted there are AI tools compatible with Australian privacy laws that could still aid in identifying victims.
“We need to look at how law enforcement can use AI legally and effectively to protect children,” he said. “There are tools available that come with proper oversight.”
Clearview AI is still in use overseas, but Gannon emphasized that Australia should adopt similar capabilities while maintaining strong oversight and collaborating internationally to ensure AI safety standards are consistent.
Experts also warned that unregulated AI is allowing child sex offenders to expand their crimes rapidly. Tame noted that even general-purpose AI chatbots are being misused by predators to automate grooming and learn how to avoid detection.
“In my own case, the man who abused me checked into a psychiatric ward as soon as he was suspended from my school,” she said. “Now we’re seeing offenders not only improve their tactics, but also their ability to escape justice.”
The government has acknowledged that current regulations don’t adequately address AI risks and has pledged to consider mandatory safety requirements. Meanwhile, the eSafety Commissioner recently reminded technology companies of their responsibility to protect children:
“While perpetrators bear the main responsibility, tech companies must be held accountable for how their tools are used.”
