Six years jail time under new Australians laws for AI porn spreaders

Australians convicted of spreading AI-generated sexually explicit images will face up to six years in prison as part of efforts to counter a spike in online exploitation and violence against women.

But experts warn a lack of awareness or understanding of technology created forms of abuse could lead to high rates of young people being penalised.

The federal government will introduce new laws to ban the sharing of AI-generated sexually explicit images – known as deepfakes – following a special national cabinet meeting last Wednesday.

It’s understood legislation will be introduced later this year and will carry a maximum penalty of six years imprisonment.

Professor Nicola Henry, a socio-legal scholar at RMIT University in Melbourne, supported the idea of having a specific criminal offence for deepfakes but raised concerns that children who didn’t fully understand the issue would be made vulnerable under the crackdown.  

“I’m thinking about a group of school kids who might think it could be funny and don’t quite realise the harms or legal implications of their behaviour,” Dr Henry said.

“I feel like it’s more important to invest resources and money into education and also we need to hold accountable the social media companies for the content to be on their platform.”

According to US-based online security researchers Home Security Heroes, there were an estimated 95,820 deepfake videos online in 2023 – 550 per cent more than in 2019.

US analysis company Graphika identified a 2000 per cent spike in the sharing of deepfakes on platforms including Reddit and X at the end of the year.

Professor Henry ran a survey last year of people who have had fake images shared and found that Australia had far higher rates of victims (3.7 per cent) compared with the global average of 2.2 per cent.

Interestingly, her research also found that men were more likely to report being perpetrators as well as victims of deepfake pornography.

“Unfortunately, there’s not a huge amount of research that’s been done to examine the extent or the impacts of deepfakes or AI-generated image-based abuse,” Professor Henry added.

“I do think it’s important to have laws in place, but it’s not going to solve this.”

The government will also bring forward legislation in early August to outlaw the release of private information online with an intent to cause harm, known as doxxing.

Attorney-General Mark Dreyfus said digitally created and altered sexually explicit material was a “deeply distressing” form of abuse, particularly against women and girls.

Domestic violence helplines