Submission 241
Dr Cassandra Cross
Submitter group: Individuals.
Position labels are not shown for individual submitters.
An AI model identified the submitter group and wrote the summary, key points and recommendation summaries. Counts do not measure importance or merit.
Summary
Read the original submission for the full text.
Argues from fraud and victimisation research that AI acts as a force multiplier for scams and psychological manipulation. It asks for a shared responsibility approach across the ecosystem, revised prevention messaging, a situational view of vulnerability and recognition of non-financial harms.
Key points
Quotes are the submitter’s own words. Text taken from documents, including scanned pages read by text-recognition software, can differ from the PDF. Check the original. Listed in the same order as on the submission’s page in the interactive guide.
Argues that AI acts as a force multiplier for existing forms of offending rather than creating entirely new categories of harm.
AI should not be viewed as creating entirely new categories of harm. Rather, it acts as a force multiplier, increasing the scale, speed, sophistication and accessibility of existing forms of offending.
Argues that the burden of protection cannot rest solely with individuals and that Australia should adopt a shared responsibility framework placing obligations across the ecosystem, including developers, platforms, telecommunications providers, financial institutions, marketplaces and government agencies.
The burden of protection cannot rest solely with individuals. Rather, Australia should adopt a shared responsibility framework that places obligations across the ecosystem, including AI developers, digital platforms, telecommunications providers, financial institutions, online marketplaces, and government agencies.
Argues that the main deepfake concern is erosion of trust in digital communications, not only fake content.
Public discussion often frames deepfakes as a problem of fake videos or fabricated images. While these concerns are valid, the broader issue is the erosion of trust across digital communications.
Argues that vulnerability is situational and can affect anyone, not only defined demographic groups.
I have argued consistently that all individuals have a potential weakness and are vulnerable to fraud, if targeted in the right way at the right time.
Argues that financial loss alone understates the harms of victimisation.
Focusing solely on the financial value of any fraud incident significantly understates the true impacts of victimisation.
Recommendations
Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.
Adopt a shared responsibility framework placing obligations across the ecosystem, including AI developers, digital platforms, telecommunications providers, financial institutions, online marketplaces and government agencies.
Australia should adopt a shared responsibility framework that places obligations across the ecosystem, including AI developers, digital platforms, telecommunications providers, financial institutions, online marketplaces, and government agencies.
Go further than the Scams Prevention Framework, with proactive system-level protections as AI becomes more capable as a tool for deception.
The introduction of the Scams Prevention Framework is a positive step which seeks to improve the system response to scams. However, this needs to go further, as the more capable AI becomes as a tool for deception, the stronger the case for proactive system-level protections.
Focus any future response on how sectors can work together to prevent, detect and disrupt AI-enabled fraud before it reaches potential victims, with responsibility also on organisations that develop, facilitate and benefit from the technologies.
responsibility must also sit with those organisations that develop, facilitate and benefit from these technologies. Any future response should focus on how different sectors can work together to prevent, detect and disrupt AI-enabled fraud before it reaches potential victims.
Rethink prevention messaging: move beyond simplistic awareness messages to consider the nature of the interaction and the practical safeguards that can be implemented across the broader system.
Future prevention efforts should recognise this changing environment and move beyond simplistic awareness messages to consider the nature of the interaction and what practical safeguards can be implemented across the broader system.
Focus discussions of AI not only on authenticity and misinformation but also on how the technologies enable highly targeted persuasion, coercion and exploitation.
Discussions around AI should therefore focus not only on authenticity and misinformation, but also on the ways these technologies enable highly targeted forms of persuasion, coercion and exploitation.
Avoid defining vulnerability only by demographic groups; focus policy on protecting people when situational vulnerability occurs.
Policy responses should be cautious about treating vulnerability as something possessed only by particular groups of people who possess prescribed demographics or a shared set of characteristics. My research consistently demonstrates that vulnerability is often situational and can affect anyone under the right circumstances. AI may increase an offender's ability to identify and exploit these periods of vulnerability. Future policy should therefore focus less on who might be vulnerable and more on how systems can better protect individuals when vulnerability inevitably occurs.
Assess emotional, psychological, physical and social consequences of AI-enabled fraud, avoiding financial loss as the primary measure of seriousness.
Financial losses provide only a partial understanding of the harm associated with fraud and scams. Victims frequently experience significant emotional, psychological, physical and social consequences that can continue long after any financial issues have been addressed. These impacts are likely to become more pronounced as AI enables more sophisticated and convincing forms of deception. Any assessment of AI- related harms should therefore recognise the full range of consequences experienced by victims and avoid treating financial loss as the primary measure of seriousness.
Ensure future policy responses do not reinforce victim-blaming narratives and recognise the skill, sophistication and adaptability of offenders when considering responsibility for harm.
Future policy responses should be careful not to reinforce victim-blaming narratives and should recognise the skill, sophistication and adaptability of offenders when considering responsibility for harm.
Support further research into AI-enabled fraud and scams, particularly the experiences of people directly affected.
there is a clear need for further research into AI-enabled fraud and scams, particularly research that examines the experiences of those directly affected.
Ground policy in victims’ encounters, responses, harms and recovery/support challenges; use ongoing lived-experience research to identify emerging risks and evaluate responses.
Effective policy requires an understanding of how individuals encounter AI-enabled deception, how they respond, the harms they experience, and the challenges they face in seeking assistance and recovery. As AI continues to evolve, ongoing research grounded in lived experience will be essential to identifying emerging risks, evaluating existing responses, and ensuring that policy remains focused on the realities of victimisation rather than assumptions about it.