This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
New Zealand's Inland Revenue department issued a warning regarding a scam campaign that impersonated Commissioner Peter Mersi and used fraudulent department branding. The scammers utilized AI-generated images and messaging to promote a fake crypto tax webinar, aiming to deceive the public into disclosing personal information. This incident is part of a broader trend of AI-enabled investment scams that have contributed to significant financial losses in New Zealand.
New Zealand Inland Revenue warned that scammers used a purported AI-generated likeness of Commissioner Peter Mersi with alleged false Inland Revenue branding, as well as misleading social media ads, to promote an allegedly fake webinar on crypto tax changes. The campaign allegedly sought to trick people into disclosing personal information.
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 campaign specifically targeted the New Zealand public to trick them into disclosing sensitive personal information for identity theft.
The scammers used AI-generated likenesses and messages to impersonate a public official and run a fraudulent crypto tax scam to steal personal information.
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 scammers intentionally deploying AI-generated content to defraud the public.
Added categories peak at 3: Substantial (Epistemic Harm, Inferred)
Scammers used AI-generated images to impersonate New Zealand's Inland Revenue Commissioner in a fraudulent crypto tax campaign aimed at identity theft. This incident represents a minor national security threat, reflecting an evolved capability in financial fraud and public official impersonation, but without systemic impacts on critical infrastructure, sovereignty, or state stability.
The AI was used as a tool by scammers to generate static assets, specifically false images and messaging, to facilitate their impersonation campaign.
Limited Risk: The report describes the use of 'artificially generated likenesses' and 'AI generated images', which fall under the category of AI-generated content such as deepfakes.
The scammers used AI to generate false images and artificially generated likenesses of the Inland Revenue Commissioner to deceive social media users.
The report clearly outlines the nature of the scam, the impersonation of the Commissioner, and the use of AI-generated content. While the exact AI model is unknown, the classification of the risk domain and behavior is highly straightforward based on the provided text.
None
Technology
Generative AI Impersonation
AI tools enable easy creation of fake images and deepfakes of public figures.
Algorithmic Targeting
Algorithms push scam ads to users based on their search history.
Weak Automated Detection
Platform moderation fails to detect slightly altered re-uploaded ads.
Data Inputs
Public Likeness Exploitation
Scammers harvest public photos and logos to train and generate fake ads.
Search History Data Profiling
Platforms feed sensitive search data into recommendation algorithms.
Human Factors
Trust in Authority Figures
Users trust messages impersonating well-known figures like the Commissioner.
Urgency of Tax Deadlines
Scammers exploit taxpayer anxiety during critical tax periods.
Process and Methods
Reactive Take-Down Process
Platforms only remove malicious ads after receiving external notifications.
Inadequate Re-upload Blocking
Slightly altered scam ads reappear the next day after being taken down.
Management
Under-resourced Ad Moderation
Insufficient investment in tools to permanently block recurring scams.
Regulatory Environment
Weak Platform Liability
Lack of strict regulations holding platforms liable for hosting scam ads.
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.
