Senior Minister of Singapore Lee Hsien Loong Allegedly Misrepresented in Two Deepfake Videos on Foreign Relations

Two deepfake videos falsely depicted Senior Minister of Singapore Lee Hsien Loong commenting on US-China relations and the South China Sea. The videos misleadingly attributed views to him and the Singapore government. Posted on TikTok, the videos amassed over 190,000 views before being removed.

Risk classification

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

The primary incident involves deepfake videos of a senior government official discussing sensitive international relations, which was intended to misrepresent state positions and influence public opinion at scale.

Additional risk subdomains

• 4.3 Fraud, scams, and targeted manipulation: The report also describes a prior deepfake incident where the minister's likeness was used to promote a fraudulent investment scam for financial gain.

Causal factors

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

The incident was caused by human malicious actors intentionally deploying deepfake generation tools to create misleading videos of a public official.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes the creation of deepfakes, which are AI-generated content subject to specific transparency obligations under the EU AI Act.

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 actors impersonating Lee Hsien Loong
• Alleged harmed parties: Singapore-United States relations, Singapore-China relations, Lee Hsien Loong, Government of Singapore, Diplomacy

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

Malicious content

Reported: The report explicitly describes malicious deepfake videos created and spread on social media platforms.

Directly caused: Two deepfake videos falsely depicting the Senior Minister were uploaded to TikTok, accumulating over 146,000 and 190,000 views respectively.

Epistemic

Reported: The report explicitly describes epistemic harm through the generation and spread of realistic AI-generated deepfakes.

Directly caused: Two deepfake videos on TikTok falsely depicted the Senior Minister discussing international relations, misleading viewers who might assume the videos are real.

Inferred additional harm: The proliferation of realistic deepfakes of political leaders degrades the shared information ecosystem and undermines public consensus reality.

People affected

• Occurrences reported: 2
• People reportedly exposed: 336000

Potential causes

Management

• Reactive Content Moderation: Platform relied on post-upload reports and queries to take action.

Technology

• Advanced Deepfake Generation: AI techniques allow realistic manipulation of visual and audio content.
• Evolving Manipulation Quality: Rapidly evolving technology makes deepfakes increasingly harder to detect.

Data Inputs

• Use of Authentic Video Footage: Real speeches and interviews were used as base data to train/alter videos.
• Lack of Content Watermarking: No robust digital watermarks to verify authentic PMO videos.

Human Factors

• Public Sharing of Deepfakes: Users sharing fake videos to warn others inadvertently increases their reach.
• Difficulty in Visual Detection: Atypical facial movements and edits are hard for average viewers to spot.

Process and Methods

• Delayed Platform Detection: Platform removed videos only after they gained significant views.
• Inadequate Proactive Filtering: Lack of automated real-time prevention of high-profile political deepfakes.

Information quality

• Classification confidence: High
• Reason for confidence: The report provides clear details about the deepfake videos, their content, view counts, and the platforms involved. The role of AI in generating these deepfakes is explicitly stated, and the response from the platform is documented.
• Ambiguities identified: The specific AI models or software used to generate the deepfakes are not identified.

Malicious actors deployed deepfake videos of Singapore Senior Minister Lee Hsien Loong on TikTok to misrepresent official government positions on sensitive geopolitical issues and promote fraudulent investment schemes. While the videos were quickly removed and caused minor immediate national security impact, they underscore the ongoing strategic challenge of synthetic media targeting state leadership and public trust.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: Singapore
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. The immediate threat has been mitigated by content removal, but the underlying capability represents an ongoing strategic concern for public communication security.
• Autonomy: Human-controlled. The AI systems were used as tools by human creators who intentionally generated and uploaded the manipulated media.
• Novelty: Established threat. Deepfakes targeting political figures and using their likeness for financial scams or political misrepresentation are an established and recurring threat.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or human safety were impacted or threatened by the deepfake videos.
• Information security: Minor. Deepfake videos manipulated authentic footage of a senior minister to falsely project government positions on sensitive geopolitical issues, but were quickly removed with limited systemic impact.
• Sovereignty: Minor. The incident targeted the representation of government positions on international relations, but did not disrupt core government operations, electoral systems, or state authority.
• Economic security: Negligible. While a prior deepfake promoted a fraudulent investment platform, there was no threat to Singapore's financial systems, strategic industries, or technological competitive advantage.
• Societal stability: Minor. The proliferation of realistic deepfakes of political leaders poses a minor risk of eroding public trust in official communications and degrading the shared information ecosystem.