This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Deepfake videos featuring the likeness and synthetic voice of Australian Prime Minister Anthony Albanese were used in scam advertisements on YouTube. The ads promoted a fraudulent cryptocurrency trading platform called AUFIRST, soliciting $400 investments from users under the guise of a government tax dividend. This incident highlights ongoing challenges with AI-generated scams on major digital advertising platforms.
Purported deepfake videos of Australian Prime Minister Anthony Albanese, featuring a synthetic voice with an American accent, were reportedly used in scam advertisements on YouTube in mid-2025. They reportedly claimed that Australians could earn thousands per month through an automated cryptocurrency trading platform called AUFIRST, framed as a government "tax dividend." Viewers were reportedly directed to a fraudulent site and urged to invest $400.
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 scam ads generated and spread false information regarding government tax dividends and investment returns to mislead the public.
The incident involves the use of AI-generated deepfakes of a public figure to run fraudulent advertisement campaigns aimed at scamming individuals for financial gain.
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 human malicious actors intentionally deploying AI-generated deepfakes to defraud the public post-deployment.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
Malicious actors utilized AI-generated deepfakes of Australian Prime Minister Anthony Albanese in fraudulent YouTube advertisements to promote a cryptocurrency scam. While representing a minor threat to information integrity and public trust in official figures, the incident is a localized, financially motivated fraud campaign with limited national security implications.
The AI was used as a tool to generate synthetic video and voice clones of public figures. The report describes the technology as being used to 'automate' or generate content, acting on direct inputs to produce specific outputs without independent planning or autonomous decision-making.
Risk Level 3: Limited Risk. The report describes 'AI-generated content such as deepfakes' which pose a moderate risk and require transparency obligations under the EU AI Act.
The primary technology used was deepfake video generation to clone the likeness of Anthony Albanese, combined with synthetic voice generation to mimic his speech.
The report clearly describes the nature of the scam, the platform used, the target audience, and the specific public figures impersonated. There is little ambiguity regarding the malicious intent or the mechanism of the scam.
None
Technology
Advanced Deepfake Technology
AI allowed highly realistic cloning of voices and likenesses for scams.
Automated Ad Delivery Systems
Fully automated systems allowed deepfake ads to bypass initial filters.
Inadequate Detection Algorithms
Detection algorithms failed to flag the synthetic Albanese deepfake.
Data Inputs
Misleading Video and Audio Inputs
Scammers used high-quality likeness and voice data of public figures.
Fictional Transcript Data
Scam used fake transcripts promising high dividends to trick users.
Unauthorized Likeness Data
Scammers exploited public audio and video to train AI clone models.
Human Factors
User Trust in Public Figures
Targeted Australians trusted the deepfake of the Prime Minister.
Sophisticated Scammer Tactics
Scammers exploited the growing trend of watching YouTube on TV screens.
Process and Methods
Ineffective Ad Review Processes
Automated and human reviews failed to block the fraudulent deepfake ad.
Flawed Ad Rejection Logic
Legitimate ads get rejected while harmful deepfakes slip through.
Management
Underinvestment in Review Controls
Failure to dedicate enough computing power and review staff to block scams.
Reliance on Fully Automated Systems
Management relied on automation, failing to protect users from fraud.
Regulatory Environment
Lack of Immediate Platform Liability
Digital platforms face fewer immediate lawsuits than traditional TV stations.
Absence of Finalised Codes
Government framework codes for digital platforms are still being finalised.
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.
