Social Media Scammers Used Deepfakes of Taylor Swift and Several Other Celebrities in Fraudulent Le Creuset Cookware Giveaways

Scammers reportedly made deepfakes of Taylor Swift, Selena Gomez, Joanna Gaines, Lainey Wilson, Ree Drummond, Oprah, Jennifer Lopez, Trisha Yearwood, Martha Stewart, and Blake Shelton promoting a Le Creuset giveaway. These AI-generated ads, appearing on Meta and TikTok, falsely claimed users could receive free cookware by paying a small shipping fee. Victims were unknowingly enrolled in a costly monthly subscription.

Risk classification

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

The incident involved scammers using AI voice-cloning technology to impersonate trusted celebrities for the purpose of defrauding social media users of their money.

Additional risk subdomains

• 6.3 Economic and cultural devaluation of human effort: The unauthorized replication and exploitation of the celebrities' public likenesses and voices devalued their personal brands and control over their own likeness.

Causal factors

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

The incident was caused by malicious human actors who intentionally deployed AI voice-cloning tools to perpetrate a financial scam.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes the use of AI to generate deepfakes, which falls under the category of AI-generated content requiring specific transparency obligations under the Act.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Voice Generation; Deepfake Video Generation
• Behaviour type: Tool
• Alleged developer: Unknown deepfake technology developers
• Alleged deployer: Unknown scammers
• Alleged harmed parties: Trisha Yearwood, Taylor Swift, Selena Gomez, Ree Drummond, Oprah, Martha Stewart, Le Creuset, Lainey Wilson, Joanna Gaines, Jennifer Lopez, General public, Fans, Blake Shelton

Harm severity

Highest direct severity in any category: Substantial. 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 Minor
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Substantial
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: Yes, the reports describe the creation and distribution of deceptive deepfake videos and cloned audio of celebrities to facilitate a scam.

Directly caused: Scammers generated and uploaded deepfake videos and cloned audio of Taylor Swift and other public figures to social media platforms.

Indirectly caused: N/A

Inferred additional harm: It is highly likely that numerous other variations of these celebrity deepfake ads were generated and distributed using similar accessible AI tools.

Privacy

Reported: Yes, the reports describe users being tricked into providing sensitive personal and financial information.

Directly caused: N/A

Indirectly caused: Victims submitted their full names, physical addresses, email addresses, phone numbers, and credit card details to fraudulent websites.

Inferred additional harm: The harvested personal and financial data of hundreds of victims could be sold on the dark web or used for further identity theft and financial fraud.

Epistemic

Reported: Yes, the reports describe the fabrication of celebrity endorsements and the creation of fake news websites to deceive users.

Directly caused: AI voice cloning was used to fabricate endorsements from Taylor Swift, Selena Gomez, and Ree Drummond, misleading fans into believing the giveaways were legitimate.

Indirectly caused: The scam eroded public trust in online advertisements, celebrity endorsements, and digital media authenticity.

Inferred additional harm: The widespread proliferation of such highly convincing deepfakes contributes to a broader erosion of shared reality and trust in digital information ecosystems.

People affected

• Occurrences reported: 1
• People reportedly harmed: 10
• People reportedly exposed: 2300

Potential causes

Management

• Inadequate Ad Pre-Screening: Platforms allowed unverified advertisers to run celebrity-themed ads.
• Lack of Endorsement Verification: Platforms failed to verify contractual endorsement rights before hosting ads.
• Delayed Ad Takedown Response: Scam ads accumulated thousands of views before being removed by platforms.

Technology

• Accessible Text-to-Speech Tools: Free and cheap tools allow anyone to clone realistic celebrity voices quickly.
• Advanced Lip-Syncing Software: Programs seamlessly sync cloned audio with existing video footage.
• Content Cloaking Techniques: Scammers cloak ad content to evade automated platform detection systems.

Data Inputs

• Abundant Public Voice Data: Large amounts of celebrity audio online are used to train voice clones.
• Stolen Brand Assets: Scammers used legitimate logos and media to build highly convincing fake sites.
• Celebrity Training Footage: Existing videos of celebrities are used as templates for deepfake overlays.

Human Factors

• Parasocial Fan Trust: Fans trust celebrity likenesses, making them vulnerable to deceptive ads.
• Inability to Spot Deepfakes: Highly realistic AI makes it nearly impossible for users to detect fraud.
• Urgency-Driven Decision Making: Fake limited-supply warnings pressure users into paying without verification.

Process and Methods

• Evaded Platform Review Systems: Social media review processes failed to catch cloaked fraudulent ads.
• Abuse of Account Verification: Verified accounts and sponsored labels lent fake legitimacy to scam ads.
• Deceptive Subscription Billing: Hidden terms and recurring fees were buried in fine print on external sites.

Regulatory Environment

• No Federal AI Scam Laws: Absence of federal legislation limits legal action against deepfake creators.
• International Jurisdictions: Scammers operating from foreign countries evade domestic law enforcement.
• Inconsistent State Laws: Only a few states regulate AI-generated content, leaving regulatory gaps.

Information quality

• Classification confidence: High
• Reason for confidence: The reports are highly consistent, originate from reputable news outlets, and clearly detail the mechanics of the scam, the platforms involved, and the impact on victims. There is minimal ambiguity regarding the role of AI or the nature of the harm.
• Ambiguities identified: The exact number of victims and the total financial losses are not fully quantified in the reports.

This incident involved a widespread consumer scam utilizing AI-generated celebrity deepfakes to perpetrate financial fraud. While it highlights the growing sophistication of synthetic media, it represents a commercial crime and consumer protection issue rather than a threat to national security, resulting in negligible national security impact.

• Overall national security impact: Negligible
• Response level: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. The incident represents a strategic, ongoing concern regarding consumer fraud rather than an active national security crisis requiring immediate response.
• Autonomy: Human-controlled. The AI voice-cloning and video tools were directly operated by human scammers to generate specific content, with no autonomous decision-making by the AI.
• Novelty: Evolved capability. The scam represents an evolution in the quality, ease of access, and scale of AI-generated deepfakes used to deceive the public.

Impact by dimension

• Physical security: Negligible. The incident involved a consumer financial scam and did not target physical systems, critical infrastructure, or human safety.
• Information security: Negligible. While deepfakes were used, this was a commercial fraud campaign targeting consumers rather than a state-sponsored information warfare operation or intelligence compromise.
• Sovereignty: Negligible. No impact on government decision-making, electoral systems, or state authority was reported.
• Economic security: Negligible. The financial losses were limited to individual consumers scammed through credit card fraud, with no systemic impact on national economic or technological security.
• Societal stability: Negligible. The incident represents a consumer protection issue rather than a large-scale threat to social cohesion, civil liberties, or human rights.