Deepfakes

Consultation has concluded. Below is a record of the engagement.



Share your views on deepfakes and help shape SA legislation.

What's being decided?

We are seeking your feedback on deepfakes.

Non-consensual deepfakes can cause significant emotional distress, reputational and economic loss. And although they’re becoming more common, their creation and use remains largely unregulated in Australia.

We’re asking all interested South Australians to share their views:

  • Are current laws suitable?
  • Do we need to reform state legislation?
  • Do you have a preferred solution for reform?

Background

Deepfakes refer to images, video or audio of a real person that have been edited, usually with Artificial Intelligence (AI) technology, to create an extremely realistic but false depiction of them doing or saying something that they did not actually do or say. It can also include wholly computer-generated images of humans that do not exist in real life. Recent advances in AI capability make this media easier to create than ever before. They’re also almost impossible to distinguish from the real thing.

Concerns are being raised about the misuse of deepfake technology to:

  • Create child exploitation material
  • Create non-consensual pornographic material (including 'revenge porn')
  • Start malicious hoaxes
  • Bully
  • Commit fraud
  • Spread misinformation
  • Manipulate public opinion.

There is growing awareness and recognition of the harm that the creation and distribution of non-consensual deepfakes can have.

We want to hear your feedback to help ensure our laws are fit for purpose, keeping our community safe online.

Separately, the Government has announced it will introduce new laws to criminalise the creation and distribution of sexually explicit deepfakes.

Get involved

Share your feedback and views on potential legislative changes to address deepfakes in South Australia.

Find out more:

Have your say by:

Please note that we may make submissions (including name and address details) publicly available. If you do not wish for this to happen, mark your submission ‘Confidential - Not for Publication’ and provide your reasons for this.

Please be aware that unless you make a request for confidentiality, we may refer to or publish the information contained in your submission. Any material identified as ‘confidential’ is still subject to the Freedom of Information Act 1991 and, while we will make an effort to keep the material confidential, in some circumstances we might disclose it under that Act. Where disclosure of information may identify you, we will attempt to consult with you under the Freedom of Information Act 1991 before the documents are disclosed.

What are the next steps?

Your feedback on how we could regulate deepfakes will be considered and a report will be provided to the Attorney-General.

We will publish a feedback summary on this website.

Share your views on deepfakes and help shape SA legislation.

What's being decided?

We are seeking your feedback on deepfakes.

Non-consensual deepfakes can cause significant emotional distress, reputational and economic loss. And although they’re becoming more common, their creation and use remains largely unregulated in Australia.

We’re asking all interested South Australians to share their views:

  • Are current laws suitable?
  • Do we need to reform state legislation?
  • Do you have a preferred solution for reform?

Background

Deepfakes refer to images, video or audio of a real person that have been edited, usually with Artificial Intelligence (AI) technology, to create an extremely realistic but false depiction of them doing or saying something that they did not actually do or say. It can also include wholly computer-generated images of humans that do not exist in real life. Recent advances in AI capability make this media easier to create than ever before. They’re also almost impossible to distinguish from the real thing.

Concerns are being raised about the misuse of deepfake technology to:

  • Create child exploitation material
  • Create non-consensual pornographic material (including 'revenge porn')
  • Start malicious hoaxes
  • Bully
  • Commit fraud
  • Spread misinformation
  • Manipulate public opinion.

There is growing awareness and recognition of the harm that the creation and distribution of non-consensual deepfakes can have.

We want to hear your feedback to help ensure our laws are fit for purpose, keeping our community safe online.

Separately, the Government has announced it will introduce new laws to criminalise the creation and distribution of sexually explicit deepfakes.

Get involved

Share your feedback and views on potential legislative changes to address deepfakes in South Australia.

Find out more:

Have your say by:

Please note that we may make submissions (including name and address details) publicly available. If you do not wish for this to happen, mark your submission ‘Confidential - Not for Publication’ and provide your reasons for this.

Please be aware that unless you make a request for confidentiality, we may refer to or publish the information contained in your submission. Any material identified as ‘confidential’ is still subject to the Freedom of Information Act 1991 and, while we will make an effort to keep the material confidential, in some circumstances we might disclose it under that Act. Where disclosure of information may identify you, we will attempt to consult with you under the Freedom of Information Act 1991 before the documents are disclosed.

What are the next steps?

Your feedback on how we could regulate deepfakes will be considered and a report will be provided to the Attorney-General.

We will publish a feedback summary on this website.

Submissions

Please note that any submissions lodged via the guestbook will be moderated prior to publication. 

If they are published, they will be visible to other participants. 

Alternatively, you can make a submission via email: LLPsubmissions@sa.gov.au

Consultation has concluded. Below is a record of the engagement.



You need to be signed in to comment in this Guest Book. Click here to Sign In or Register to get involved

Yes, we need to reform state legislation. Current laws need to be more explicit to include AI, deepfakes and technology abuse and include anticipated hypothetical and actual scenarios. Not all of the following points relate to legislative change requirements, but need consideration:
- public education is required on how victims can seek assistance (which may also include counselling, coping mechanisms, resilience)
- education of consumers of deep fake images i.e. what is driving people to look at pornographic and child abuse material and what is their understanding of the harm created. What counselling or personal change support is available?
- requirement to label all deepfake images or visa versa i.e. labelling of genuine images (although this would be difficult to police). Identification and reporting could become a community responsibility
- can digital images ever be fully erased once they become public? i.e. a victim may never be free of the threat, especially children who may have a lifetime to live with the knowledge their image is "out there". Mental harm may restrict someone's future & their earning capacity. Penalties need to reflect this.
- compensation to victims directly from perpetrators (which may include access to their superannuation and other assets, time in community service, etc)
- use of AI to restrict the movements of perpetrators in the community (we don't need more people in jails) or to "digitally" mark them in a way that the general community can be aware of the crime (a form of public shaming which would be experienced by victims).
- perhaps the burden of proof in current legislation is too high?
- there has been a fantastic amount of information provided to the public regarding scammers, scam alerts, etc. I assume this will continue and include deepfake scams as well.
Thank you for the opportunity to comment, even when we are all still learning and may have unrealistic expectations about what can be achieved by Government. The more we can get citizens involved in resolving issues the better.

Heather Falckh 29 days ago

Are current laws suitable? As with any technology, policy or views, legislation need to continue to change and improve to keep up with the times. A review and improvement of the law to cover deepfakes is much needed.

Do we need to reform state legislation?
Yes, see above answer. Any means necessary is required to cover the harm caused by deepfakes and increasing use of similar technology in a maniacal manner.

Do you have a preferred solution for reform?
We focus a lot on the younger generation, which is great, however increasingly Gen Z and Gen Alpha are the ones who are technology literate and are empowered by technology to make informed decisions. It is the older generation - Gen X and Baby Boomers who equally deserve attention in how they interact with social media, and may not have the digital literacy to understand what is real and what is not. Younger people, especially the emerging generation, are curious and have a greater grasp of what is possible through technology. The older generation may easily accept what is out there as 'real', as seen by the increasing sharing of fake information, news and clickbait articles by older people especially on platforms like Facebook and Whatsapp. They also act as echo-chambers of misinformation where it is a case of "if my peers acknowledge and think this is real, it must be", but lack the literacy to tell what is real and what are deepfakes. In addition, and probably more concerning, is that deepfakes are increasingly used by right-wing or those with extremist views wanting to influence all sections of the population to harmful discourse, not just towards impressionable teens or kids. Targeting these cohorts should exist alongside young people.
Overall I think it is great that the government is addressing this and I look forward to seeing a more digital savvy Australia who can think more critically between actual facts and malicious 'truths'.

Daryl Tian about 1 month ago

Are current laws suitable? No, there needs to be new and updated laws to cover and keep up with this type of technology, which can be detrimentally harmful to citizens.
Do we need to reform state legislation? Yes, although many laws that can come into effect regarding deep fakes, people need an easy and accessible way to manage harmful situations with deepfakes, in conjunction with our legal system, to navigate through what can be an incredible stressful time for adults and minors if affected by the technology.
Do you have a preferred solution for reform? Education campaigns, the monitoring of creators of deepfakes with illegal content, support systems in place for victims of the technology, further laws to establish grounds for managing social repercussions that stem from misinformation through deepfakes.

EmmaDig about 1 month ago

I support the full force of the law being enabled in SA to prosecute perpetrators of deep fake technology and/or revenge porn. I support all legislative reforms being considered, as well as establishment of a compensation fund or access to legal support in the event of a victim's need to prosecute. Police also must be on the same page, enabling smooth access to due process when instances are reported. All government bodies need to be aligned to the same purpose. In addition, I would like the Parliament to consider public education campaigns and school education programs aimed at high school students to further embed critical thinking and questioning of dubious images circulated in the public domain. Age-appropriate education should also start in primary schools, to build the next generation of questioning young people who don't take visual imagery at face value.

Anne Marie Taplin about 1 month ago

Deepfakes absolutely need to be policed to prevent misinformation. As a society we must protect younger children from falling for this misinformation as well.

Phil R about 1 month ago