Deepfake Video Reportedly Depicts U.S. Congressman Rob Wittman Endorsing Military Support for Taiwan's Democratic Progressive Party

A deepfake video is reported to have falsely depicted U.S. Congressman Rob Wittman endorsing military support for Taiwan’s Democratic Progressive Party candidates in the 2024 presidential election. Shared on TikTok, the video is reported to have undermined Taiwanese voter confidence in the DPP by alleging U.S. interference. Fact-checking from AFP finds Wittman never made the statements attributed to him.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.1 Disinformation, surveillance, and influence at scale

The deepfake video was deployed as part of a coordinated effort to influence public opinion and manipulate the political process of the Taiwanese presidential election.

Causal factors

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

The incident was caused by human actors who intentionally created and distributed a deepfake video post-deployment to manipulate public opinion.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes the creation of a deepfake video. Under the EU AI Act, AI-generated content such as deepfakes falls under Limited Risk, which requires specific transparency obligations to ensure users are informed.

AI system and alleged parties

• AI system: None named
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Unknown deepfake technology developers
• Alleged deployer: Unknown actors in China
• Alleged harmed parties: Taiwanese voters, Rob Wittman, Democratic Progressive Party

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 Minor
• 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 Minor, indirect Minor
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: Yes, the report describes the creation and spread of a deepfake video intended to mislead voters.

Directly caused: The deepfake video falsely depicted US Representative Rob Wittman endorsing military support for Taiwan's DPP.

Indirectly caused: The video was shared on TikTok with traditional Chinese captions to stoke false voter concerns about US meddling.

Inferred additional harm: It is likely that the video was shared across other social media platforms, exposing more voters to the malicious content.

Democracy

Reported: Yes, the report describes an attempt to influence the Taiwanese presidential election using deepfake technology.

Directly caused: The deepfake video targeted the January 13 presidential election in Taiwan to undermine support for the DPP.

Indirectly caused: The video aimed to stoke false voter concerns about US meddling in the election, potentially manipulating democratic outcomes.

Inferred additional harm: Such incidents erode public trust in democratic institutions and the integrity of elections.

Epistemic

Reported: Yes, the report describes the fabrication of a video interview to spread false claims.

Directly caused: The video falsely portrayed Wittman as making statements about accelerating arms sales and training Taiwan's army.

Indirectly caused: Fact-checkers had to verify the footage against a 2022 interview to prove Wittman never mentioned Taiwan, showing the erosion of shared truth.

Inferred additional harm: The proliferation of high-quality deepfakes makes it harder for citizens to distinguish real news from fabricated media, degrading the overall information ecosystem.

People affected

• Occurrences reported: 1
• People reportedly exposed: 10000

Potential causes

Management

• Inadequate Risk Mitigation: TikTok management failed to prevent election interference via deepfakes.

Technology

• Deepfake Generation Tools: AI tools allowed realistic synthesis of Wittmans voice and video.
• Algorithmic Distribution: TikTok algorithms amplified the reach of the manipulated video.

Data Inputs

• Source Video Exploitation: A real 2022 WUSA TV interview was used to train or map the deepfake.
• Manipulated Captions: Traditional Chinese captions were added to mislead non-English speakers.

Human Factors

• Malicious Actor Intent: State-aligned actors targeted Taiwanese voters to undermine the DPP.
• Voter Susceptibility: Voters may believe realistic video evidence without checking sources.

Process and Methods

• Inadequate Platform Detection: TikTok failed to automatically detect and flag the synthetic media.
• Delayed Fact-Checking: Manual reverse image search was required to identify the original source.

Regulatory Environment

• Weak Platform Liability: Lack of strict regulations holding platforms liable for political deepfakes.

Information quality

• Classification confidence: High
• Reason for confidence: The report clearly details the nature of the deepfake, the platform it was shared on, the target of the disinformation, and the fact-checking process that debunked it. There is little ambiguity regarding the intent or the impact of the incident.
• Ambiguities identified: The specific AI tool used to generate the deepfake is not identified.

A deepfake video of US Representative Rob Wittman was deployed on TikTok to influence the 2024 Taiwanese presidential election. The incident represents a coordinated information warfare effort to manipulate democratic processes and stoke geopolitical tensions, highlighting the risks of synthetic media in international relations.

• Overall national security impact: Substantial
• Scope: Multiple nations
• Primary target: Taiwan
• Other affected: United States
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Near-term. The deepfake was released on December 29, 2023, targeting the Taiwanese presidential election scheduled for January 13, 2024.
• Autonomy: Human-controlled. The AI system acted as a tool for content generation under the direct control and intent of human actors.
• Novelty: Evolved capability. While political deepfakes are an established threat, this represents an evolved capability targeting cross-strait relations and international elections.

Impact by dimension

• Physical security: Negligible. No physical threats, kinetic attacks, or critical infrastructure compromise reported in connection with this deepfake video.
• Information security: Substantial. AI-generated deepfake targeted a foreign election by fabricating a US Representative's endorsement, representing a notable foreign influence and disinformation operation.
• Sovereignty: Substantial. The incident represents a direct attempt to manipulate the electoral process of Taiwan and falsely project US government foreign policy decisions.
• Economic security: Negligible. No strategic technology theft, financial system attacks, or significant economic impacts were reported.
• Societal stability: Minor. Potential to stoke public anxiety regarding US interference in Taiwanese elections, though managed within standard fact-checking and platform moderation procedures.