This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Greek Finance Minister Kyriakos Pierrakakis and the Ministry of Economy and Finance have filed a lawsuit against operators of a Facebook page that used AI-generated deepfakes to impersonate the minister. The deepfake video falsely depicted the minister endorsing fraudulent high-yield investment schemes. This incident is part of a broader trend of deepfake-driven financial scams in Greece targeting citizens by impersonating public figures.
Greek Finance Minister Kyriakos Pierrakakis and the Ministry of Economy and Finance reportedly filed a lawsuit against unidentified operators of a Facebook page that allegedly used AI to generate a deepfake video falsely depicting the minister endorsing fraudulent "high-yield" investment schemes. Authorities noted this follows similar deepfake-driven scams in Greece involving impersonation of public figures for illicit financial gain.
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 deepfake video generated and spread false information by depicting the minister endorsing a non-existent investment program.
The incident involves the use of AI-generated deepfakes of a public official to run a fraudulent investment scam, which directly falls under fraud, scams, and targeted manipulation.
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 actors intentionally deploying an AI-generated deepfake post-deployment to run a fraudulent investment scam.
Added categories peak at 3: Substantial (Epistemic Harm, Inferred)
Unidentified actors used AI-generated deepfakes to impersonate the Greek Finance Minister in a fraudulent Facebook investment scheme. While targeting a high-profile government official, the incident represents localized financial fraud rather than a coordinated threat to national security, sovereignty, or critical infrastructure.
The AI system was used as a tool to generate a deepfake video based on direct inputs (the minister's likeness and voice) to produce a specific synthetic output. The report states the page used artificial intelligence to produce a deepfake video showing Pierrakakis endorsing a scam.
Risk Level 3: Limited Risk. The report explicitly mentions that under the EU's AI Act, 'deepfakes are categorized as limited risk', which requires transparency obligations to ensure users are informed.
The AI was used to generate a deepfake video and mimic the likeness and voice of the minister to make it appear as though he was endorsing a scam.
The report clearly outlines the nature of the incident (a deepfake scam targeting a minister), the legal action taken, and the broader context of deepfake scams in Greece. The roles of the actors and the AI's involvement are unambiguous.
None. The incident is a straightforward case of deepfake-enabled financial fraud.
While the specific AI model and exact financial losses are not specified, this information is not strictly necessary to confidently classify the incident under the provided taxonomies.
Technology
AI Deepfake Generation
AI used to create a realistic fabricated video of Minister Pierrakakis.
Deceptive Ad Distribution
Facebook page used to distribute fake video as sponsored advertisements.
Data Inputs
Impersonation Data Exploitation
Exploitation of public media of the minister to generate realistic deepfake.
Human Factors
Malicious Actor Exploitation
Unidentified operators created deepfakes to defraud citizens for profit.
Process and Methods
Inadequate Ad Screening
Platform failed to detect and block AI-generated scam ads before publication.
Regulatory Environment
Limited Risk Classification
Under EU AI Act, deepfakes are classified as limited risk, limiting oversight.
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.
