Study finds one in 15 adults victimised by explicit deepfakes

Summarised from 2 outlets · 4 reports · Updated 4 Nov, 10:06 · Archive

One in 15 adults have experienced abuse involving non-consensual sexually explicit AI-generated images, according to an international study of 7,230 people.

A study by RMIT University surveyed more than 7,200 adults in Australia, the United States and the United Kingdom in July and August 2024 to investigate abuse involving non-consensual AI-generated sexually explicit images, commonly known as deepfake pornography. The research, funded by Google and published on Wednesday, found around one in 15 adults had experienced at least one form of such abuse since turning 18. This includes the creation and sharing of intimate images generated by AI without consent, as well as threats to do so. In Australia specifically, the figure was slightly higher at 7.7 per cent.

Young people aged under 35 were disproportionately affected, experiencing abuse at more than 4.35 times the rate of older groups. The study also found higher rates among marginalised racial and ethnic groups and people with disabilities. Men reported the highest rates of having explicit AI images created of them without consent, with 10.2 per cent experiencing abuse compared with 3.8 per cent of women. However, women reported a broader range of harms and greater negative emotional impacts, with women much more likely to feel depressed, sad or afraid for their safety.

Nearly 87 per cent of those who experienced such abuse reported at least one impact on their daily lives, affecting their health and wellbeing, work, relationships or finances. Lead author Nicola Henry from RMIT University said the true scale of the problem may be even greater, as some people may never know that AI-generated images of them have been created or shared.

In Australia, the federal government established nationwide criminal penalties in 2024 that made sharing non-consensual deepfake pornography online an offence. Victoria, New South Wales, South Australia, Western Australia and Queensland have made the creation of such images a specific crime, regardless of whether they are shared. Tasmania remains the only Australian jurisdiction yet to criminalise the non-consensual sharing of AI-generated explicit images, with draft legislation due for public consultation by the end of the year.

How it is being reported

• New report reveals one in 15 adults have suffered forms of AI-generated image abuseThe horrifying rate of Aussie adults who have experienced abuse from sexually explicit AI-generated images has been revealed in a new study.The West Australian · 4 Nov, 03:47
• New report reveals one in 15 adults have suffered forms of AI-generated image abuseThe horrifying rate of Aussie adults who have experienced abuse from sexually explicit AI-generated images has been revealed in a new study.Perth Now · 4 Nov, 03:47
• Young people at greatest risk from explicit 'deepfakes'Men are most likely to have explicit AI generated images created of them, but women generally report the most negative impacts, including fear for their safety.Perth Now · 3 Nov, 20:01
• Young people at greatest risk from explicit 'deepfakes'Men are most likely to have explicit AI generated images created of them, but women generally report the most negative impacts, including fear for their safety.The West Australian · 3 Nov, 20:01

In this story: RMIT University · Nicola Henry · Australia · United Kingdom · United States · Google · Tasmania · Victoria

This summary was written by AI from the headlines and standfirsts above, and states as fact only what at least two outlets report. How we use AI · Report a problem · Corrections log