This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A man was ordered by the Australian Federal Court to pay $343,500 for creating and distributing non-consensual pornographic deepfake images of prominent Australian women. Despite receiving a removal notice from the eSafety Commissioner, the individual continued to distribute the content, leading to contempt of court charges. The case also involved separate allegations regarding the distribution of deepfake images of students and teachers at a Brisbane school.
In a first-of-its-kind case under Australia's Online Safety Act, the Federal Court ordered Anthony Rotondo to pay $343,500 plus costs for creating and sharing non-consensual deepfake pornographic images of prominent Australian women on MrDeepFakes.com. After reportedly receiving a removal notice in 2023, Rotondo defied authorities, emailed the images to dozens of recipients, and was later found in contempt. The images were reportedly taken down after he provided account access to investigators.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to gain a personal advantage over others such as through cheating, fraud, scams, blackmail or targeted manipulation of beliefs or behavior. Examples include AI-facilitated plagiarism for research or education, impersonating a trusted or fake individual for illegitimate financial benefit, or creating humiliating or sexual imagery.
Also classified as
The distribution of explicit deepfake pornographic images to 50 email addresses and online platforms exposed recipients to toxic and abusive content.
The perpetrator used AI deepfake tools to create non-consensual sexually explicit and humiliating imagery targeting specific prominent women, students, and teachers.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The incident was caused by a human user intentionally using deepfake generation tools post-deployment to create and distribute non-consensual explicit images.
Added categories peak at 3: Substantial (Child Sexual Exploitation & Abuse (CSEA), Direct)
An Australian court fined an individual $343,500 for creating and distributing non-consensual explicit deepfakes of prominent women, students, and teachers. The incident highlights the growing threat of AI-enabled image abuse to societal stability and individual human rights, while demonstrating increasing regulatory and legal enforcement capabilities against such offenses.
The AI system (deepfake generator) was used as a tool by a human actor to perform a specific function (generating synthetic images from inputs) without autonomous planning or agency.
Limited Risk: The report describes the creation of AI-generated content such as deepfakes, which falls under Risk Level 3 (Limited Risk) due to transparency obligations.
The AI was used specifically to generate deepfake pornographic images of targeted individuals.
The report provides clear details about the legal actions, the perpetrator's actions, and the nature of the deepfakes. The roles of the eSafety Commissioner and the Federal Court are well-documented.
None. The incident is clearly a case of non-consensual deepfake creation and distribution.
Technology
Deepfake Generation Tools
AI tools enabled the creation of realistic non-consensual explicit images.
Data Inputs
Victim Source Images
Existing images of prominent women were used to generate the deepfakes.
Human Factors
Defiance of Authority
The perpetrator ignored eSafety notices and sent images to 50 addresses.
Malicious Intent
The perpetrator deliberately sought to harass and cause distress to victims.
Process and Methods
Ineffective Content Containment
Processes failed to prevent the perpetrator from emailing copies of images.
Regulatory Environment
Jurisdictional Gaps
Perpetrator claimed immunity due to residing outside Australia.
AI system involved
An AI system was involved, but the reports do not identify it.
Identified by automated extraction; can differ from the alleged developers above.
