Laws banning the use of Artificial Intelligence to generate violent or sexually explicit deepfake images, audio or video will be debated in State Parliament this week.
The bill represents a significant collaboration between Upper House MLC Connie Bonaros and the State Government, to ensure the laws are comprehensive, robust and tackle the emerging issue.
Under the Bonaros bill, which will be debated in the Legislative Council later today, offenders who use Artificial Intelligence (AI) to generate simulated humiliating, degrading or invasive images, audio or video purporting to be of real people could face up to four years in jail or fines of up to $20,000.
Currently, Part 5A of the Summary Offences Act already has several offences that apply to the distribution of humiliating, invasive or indecent deepfakes that have altered the image or video of a real person – however the proposed Bill will ensure that our laws also capture content that has been wholly created by AI.
Authorities estimate as much as 90-95 percent of deepfakes are non-consensual porn; and 99 per cent of victims in deepfake porn are women. The eSafety Commissioner has stated explicit deepfakes have increased on the internet as much as 550 per cent year on year since 2019.
While a Government Bill was also introduced to close this loophole, the Government was pleased to work collaboratively with the Honourable Ms Bonaros to ensure that whatever Bill passes will best achieve the outcome of protecting potential victims of these malicious and degrading attacks. Assistant Minister for Artificial Intelligence, Michael Brown MP, undertook extensive work on this issue as Chair of the Select Committee on Artificial Intelligence and was instrumental in the development of this Bill.
The Government will be supporting the Bonaros Bill in its amended form, and not progressing the Government Bill that would have had the same effect.
Once the Bill passes Parliament, the Government will work with authorities to determine how quickly these important new laws can come into effect.
As put by Kyam Maher
While this technology is arguably still in its infancy, it can already be used to disturbing effect to create offensive and often sexualised deepfake images, video and audio.
These are tough penalties that will tackle this emerging issue head on, and better protect the community.
South Australia will not tolerate the use of AI to develop sick, degrading material.
As put by Michael Brown
There is tremendous potential to use artificial intelligence in ways that benefit all South Australians.
But with those opportunities come tremendous risk, which is why it is so important that we have a solid framework in place to tackle the dark side of its use before it becomes entrenched.
The use of AI to create invasive or degrading deepfake images, audio or video is utterly vile, and these changes send a clear message that such misuse will not be tolerated in our state.
As put by Connie Bonaros MLC
For too long, our laws have lagged behind technology. Victims of deepfake abuse have been left exposed, traumatised, and without real protection.
These reforms send a powerful message: no matter how content is created, if it humiliates, degrades or invades someone's dignity without consent, it will not be tolerated.
When I first raised these concerns in Parliament, it was clear this was bigger than any single party or member. I'm proud we were able to work together to prioritise the safety and dignity of South Australians. This is about real justice for real harm. It's about giving power back to victims. And it's about ensuring our laws keep pace with the realities of the digital world.
