Purported Deepfake Video Falsely Depicts Biden Announcing National Draft for Ukraine

A purportedly AI-generated deepfake video falsely depicting President Biden announcing a national draft to support Ukraine was reportedly shared on social media, causing widespread misinformation. The video reportedly misled the public until debunked by fact-checkers.

Risk classification

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

The deepfake video of the President declaring a military draft was used to influence public opinion and political discourse at scale on social media.

Additional risk subdomains

• 3.1 False or misleading information: The AI-generated video spread false information about a national draft, leading to public confusion and misleading tweets.

Causal factors

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

The incident was caused by human actors intentionally creating and distributing a deepfake video of President Biden to influence public opinion post-deployment.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The system is used to generate AI-generated content such as deepfakes, which under the Act requires specific transparency obligations to ensure users are informed.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Unknown voice cloning technology developers, Unknown deepfake technology developers
• Alleged deployer: Jack Posobiec, @ThePatriotOasis
• Alleged harmed parties: Ukraine, Truth, Relations between the United States and Ukraine, National security and intelligence stakeholders, Joe Biden, General public of Ukraine, General public of the United States, General public, Epistemic integrity, Biden administration

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 deceptive deepfake video.

Directly caused: An AI-generated deepfake video of President Joe Biden falsely announcing a military draft was created and shared on Twitter.

Indirectly caused: The video was subsequently shared and quoted by other social media accounts, spreading the fabricated claim further.

Democracy

Reported: Yes, the report describes an incident that could erode trust in democratic institutions and leadership.

Directly caused: The deepfake falsely depicted the U.S. President invoking the Selective Service Act, fabricating an official government action.

Indirectly caused: The spread of the video without disclosure misled citizens into believing a military draft was active, potentially undermining trust in official communications.

Epistemic

Reported: Yes, the report explicitly describes the generation and spread of false information via a deepfake.

Directly caused: The AI-generated video fabricated a speech by President Biden, creating false evidence of a national draft.

Indirectly caused: Social media users were misled into believing the draft was real, leading to tweets quoting the fabricated speech as fact.

People affected

• Occurrences reported: 1

Potential causes

Management

• Prioritizing Engagement: Social media platforms prioritize engagement over strict content accuracy.

Technology

• Generative Adversarial Networks: GANs enabled highly realistic synthetic video and audio generation.
• Imperfect Deepfake Realism: Static body movement with only mouth moving indicates technology limitations.

Data Inputs

• Use of Existing Footage: Existing footage of Biden talking was manipulated to match a fake script.
• Fabricated Audio Script: Synthesized audio script was created to impersonate the President's voice.

Human Factors

• Intentional Misinformation: Creators and commentators shared deepfakes to mislead social media users.
• Lack of User Verification: Social media users shared the video without verifying its authenticity.
• Selective Context Sharing: Commentators omitted the disclaimer in tweets, presenting the video as real.

Process and Methods

• Inadequate Platform Moderation: Twitter lacked automated checks to flag or remove misleading AI deepfakes.
• Absence of Watermarking: The AI tool did not embed visible or invisible watermarks on the output.

Regulatory Environment

• Lack of Deepfake Regulations: No legal frameworks exist to restrict creating deepfakes of public officials.

Information quality

• Classification confidence: High
• Reason for confidence: The report from Snopes provides clear, verified details about the video, its creators, how it was distributed, and the fact that it was a deepfake. There is little ambiguity about the events or the technology used.
• Ambiguities identified: The specific AI tools or models used to generate the deepfake are not named in the report.

An AI-generated deepfake video of President Joe Biden falsely announcing a national military draft was circulated on social media. While the incident caused temporary public confusion and demonstrated the evolving threat of synthetic media targeting national leadership, it was quickly debunked and had minor overall impact on national security.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Domestic social media commentators

Threat characteristics

• Imminence: Long-term. The specific incident has been resolved and debunked, posing no immediate crisis, though the strategic threat of political deepfakes remains long-term.
• Autonomy: Human-controlled. The AI system was used strictly as a tool by human creators to generate the synthetic video and audio based on direct inputs.
• Novelty: Evolved capability. While deepfakes of political figures have occurred before, synthesizing a presidential announcement of a military draft represents an evolved and highly sensitive capability.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or human safety components were targeted or affected by the deepfake video.
• Information security: Minor. An AI-generated deepfake of the President falsely announced a military draft. While it targeted national leadership, it was quickly debunked by fact-checkers, representing a minor information security incident.
• Sovereignty: Minor. The video fabricated an official presidential declaration of a military draft, but it did not disrupt actual government decision-making or constitutional processes.
• Economic security: Negligible. The incident did not target financial systems, strategic technologies, or critical supply chains.
• Societal stability: Minor. The video had the potential to cause public anxiety regarding a military draft, but there is no evidence of widespread social instability or civil unrest.