Senator Ben Cardin Reportedly Received a Purported Deepfake Zoom Call Impersonating Former Ukrainian Foreign Minister Dmytro Kuleba

Senator Ben Cardin was reportedly targeted with a purported deepfake on a Zoom call that appeared to imitate former Ukrainian Foreign Minister Dmytro Kuleba. The allegedly AI-generated video reproduced his likeness and voice but drew attention when the caller asked unusual questions.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The incident involved a highly targeted impersonation of a trusted foreign official using real-time deepfake technology to manipulate and bait a specific victim, Senator Ben Cardin.

Additional risk subdomains

• 4.1 Disinformation, surveillance, and influence at scale: The operation was part of a broader, coordinated social engineering campaign by foreign actors aimed at influencing U.S. politics and elections.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The incident was caused by a human actor who intentionally deployed deepfake technology to impersonate a foreign official and deceive a U.S. Senator.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3: Limited Risk. The incident involves the use of 'AI-generated content such as deepfakes' to impersonate a foreign official. Under the EU AI Act, deepfakes are classified as limited risk, requiring transparency obligations to inform users that the content is artificially generated or manipulated.

AI system and alleged parties

• AI system: None named
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Unknown voice cloning technology developers, Unknown deepfake technology developers
• Alleged deployer: Unknown malicious actor impersonating Dmytro Kuleba
• Alleged harmed parties: Truth, Relations between the United States and Ukraine, National security and intelligence stakeholders, Epistemic integrity, Dmytro Kuleba, Ben Cardin

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Minor. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Minor, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Minor, indirect Minor
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: The report explicitly describes malicious content created and spread directly by the incident.

Directly caused: The threat actors created a real-time deepfake video and audio stream mimicking former Ukrainian Foreign Minister Dmytro Kuleba to deceive a U.S. Senator.

Democracy

Reported: The report explicitly describes concerns regarding the erosion of democratic norms caused directly and indirectly by the incident.

Directly caused: A U.S. Senator and Chairman of the Senate Foreign Relations Committee was targeted in an attempt to manipulate his foreign policy stances and bait him into election-related comments.

Indirectly caused: The incident prompted the Senate Security Office to issue warnings to all senators and staff, increasing suspicion and altering communication protocols.

Inferred additional harm: Similar deepfake operations could target other lawmakers, potentially leading to successful manipulation, leaks of sensitive national security information, or compromised diplomatic relations.

People affected

• Occurrences reported: 1
• People reportedly harmed: 1
• People reportedly exposed: 5

Potential causes

Management

• Inadequate Verification Protocols: Senate offices lacked strict pre-verification steps for virtual meetings.
• Need for Awareness Training: A lack of timely deepfake recognition training left staff vulnerable.

Technology

• Live Deepfake Piping Tools: AI tools allowed seamless integration of live video and audio deepfakes.
• Decreased Technical Barriers: Recent leaps in AI make sophisticated deepfakes easier to conduct.
• High Technical Believability: The deepfake matched the target appearance and voice with high realism.

Data Inputs

• Public Media of Target: Abundant media of Kuleba enabled highly accurate AI voice and face training.
• Spoofed Email Request: A deceptive email initiated the contact, masquerading as a trusted source.
• Relationship Intelligence: Attackers used known relationship details to make the contact look legitimate.

Human Factors

• Initial Trust in Visuals: Staff initially trusted the call because the face and voice matched Kuleba.
• Behavioral Suspicion: The senator grew suspicious due to out-of-character, aggressive questions.

Process and Methods

• Lack of Real-Time Detection: There are no foolproof methods to detect deepfakes during a live call.
• Weak Identity Verification: The office failed to verify the caller identity prior to starting the Zoom.
• Active Social Engineering: Threat actors leverage existing relationships to bypass traditional defenses.

Regulatory Environment

• Lack of AI Distribution Bans: The availability of deepfake creation tools remains largely unregulated.
• Foreign State Interference: Malign foreign actors exploit AI to target democratic processes.

Information quality

• Classification confidence: High
• Reason for confidence: The reports are highly consistent, detailed, and corroborated by official Senate security notices, Senator Cardin's own statements, and multiple major news outlets. The facts of the Zoom call, the impersonation of Dmytro Kuleba, the specific questions asked, and the subsequent security warnings are clearly documented without conflicting accounts.
• Ambiguities identified: The exact identity or state-affiliation of the 'malign actor' remains unconfirmed, with suspicions pointing to Russia, China, or Iran.
• Alternative interpretations: None. The event is clearly a targeted social engineering attempt leveraging real-time deepfake technology.

Senator Ben Cardin was targeted in a sophisticated social engineering attempt using real-time deepfake technology to impersonate former Ukrainian Foreign Minister Dmytro Kuleba. The incident represents an evolved capability in information warfare and a direct threat to government functions, prompting security warnings across Congress.

• Overall national security impact: Substantial
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Near-term. Active social engineering campaign targeting congressional offices represents a developing situation requiring monitoring and defensive adjustments.
• Autonomy: Human-controlled. The AI system functioned as a tool, mapping the human operator's inputs to the likeness of Dmytro Kuleba in real-time, with no autonomous decision-making.
• Novelty: Evolved capability. While deepfakes are established, the use of real-time, interactive video and audio deepfakes to target high-level government officials represents a significant technical evolution.

Impact by dimension

• Physical security: Negligible. No kinetic threat, critical infrastructure damage, or physical safety risk was reported in this incident.
• Information security: Substantial. Sophisticated real-time deepfake targeting a senior lawmaker to bait sensitive foreign policy responses. Promoted security warnings to Congress, showing notable intelligence security implications.
• Sovereignty: Substantial. Direct targeting of the Senate Foreign Relations Committee Chairman in an attempt to manipulate or compromise U.S. foreign policy positions and legislative communication.
• Economic security: Negligible. No direct threats to economic stability, financial systems, or strategic industrial technology were reported.
• Societal stability: Minor. While aimed at political manipulation, the incident was a highly targeted social engineering attempt rather than population-scale societal disruption.