Purported Deepfake Featuring Dr. Rinki Murphy and Jack Tame Reportedly Used to Promote Diabetes Scam in New Zealand

A purported AI-generated deepfake video reportedly impersonated Auckland University diabetes expert Dr. Rinki Murphy, depicting her in a TVNZ interview with journalist Jack Tame promoting a fake diabetes cure. The alleged scam video circulated on social media in April–June 2025, prompting concern that New Zealanders were deceived into stopping prescribed medications. Despite reporting the incident, new variants of the video reportedly continued to emerge.

Risk classification

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

The incident involves malicious actors using AI-generated deepfake videos of trusted public figures to run a targeted financial scam and manipulate vulnerable diabetes patients.

Additional risk subdomains

• 1.2 Exposure to toxic content: The deepfake video exposed users to highly unsafe medical advice, encouraging diabetes patients to stop taking their prescribed medications.
• 5.1 Overreliance and unsafe use: Vulnerable patients trusted and relied on the deepfake video's medical claims, leading them to make unsafe health decisions.

Causal factors

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

The scam was intentionally created and deployed by human malicious actors using AI deepfake technology post-deployment to defraud individuals.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3: The incident involves AI-generated content such as deepfakes, which are classified as Limited Risk under the EU AI Act, requiring transparency obligations so users know they are interacting with synthetic media.

AI system and alleged parties

• AI system: None named
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: unknown
• Alleged developer: Unknown voice cloning technology developers, Unknown deepfake technology developers
• Alleged deployer: Unknown scammers
• Alleged harmed parties: Rinki Murphy, Jack Tame, General public of New Zealand, Diabetes patients in New Zealand, Diabetes patients

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

Physical

Reported: Yes, the report explicitly states the video convinced those with diabetes to stop taking important medications, prompting a warning from Health NZ.

Indirectly caused: The video convinced diabetes patients to stop taking their medically necessary diabetes medications, posing a severe risk to their physical health.

Inferred additional harm: It is highly likely that some of the estimated 50+ affected individuals suffered adverse health outcomes or physical deterioration from discontinuing their diabetes medications, though specific numbers of hospitalizations or injuries are not detailed.

Malicious content

Reported: Yes, the report describes the creation and spread of a malicious deepfake video.

Directly caused: An AI-generated deepfake video was created impersonating a journalist and a diabetes expert to promote a fake cure.

Indirectly caused: The video was widely shared and circulated on social media platforms like Facebook, reaching an unknown but likely large audience.

Inferred additional harm: Multiple variations of the deepfake video continue to pop up on social media platforms even after being reported and taken down, expanding its reach.

Psychological

Reported: Yes, the report describes emotional distress and feelings of betrayal.

Indirectly caused: Professor Rinki felt 'horrified' that her likeness was used, her patients felt 'betrayed', and victim Steve Davis felt 'quite gutted' and described the scam as 'disgusting'.

Inferred additional harm: It is likely that many other vulnerable patients who fell for the scam experienced significant distress, anxiety, and feelings of violation upon realizing they were defrauded.

Epistemic

Reported: Yes, the report describes the spread of false medical information and the erosion of trust in public figures.

Directly caused: The deepfake video fabricated a TVNZ Q+A interview and a medical breakthrough, spreading false claims that a '5 second method' could cure diabetes.

Indirectly caused: The fabrication misled viewers into believing authentic public figures endorsed the product, eroding trust in legitimate medical professionals and journalists.

Inferred additional harm: The persistent circulation of such deepfakes online contributes to a broader erosion of shared reality and public trust in digital media and health institutions.

People affected

• Occurrences reported: 1
• People reportedly harmed: 51
• People reportedly exposed: 51

Potential causes

Management

• Prioritizing Ad Revenue: Social media platforms prioritize ad sales over strict advertiser verification.
• Slow Incident Response: Ineffective escalation paths for victims and experts reporting identity theft.

Technology

• Sophisticated Deepfake Generation: AI tools allow highly realistic cloning of voices and facial movements.
• Inadequate Algorithmic Detection: Platform algorithms fail to automatically flag and block deepfake scam ads.

Data Inputs

• Exploitation of Public Media: Scammers use public broadcast footage of journalists and medical experts.
• Lack of Media Provenance Data: No digital watermarks or metadata to verify the authenticity of the video.

Human Factors

• Trust in Public Figures: Victims trust recognizable TV presenters and medical professionals blindly.
• Desire for Chronic Illness Cure: Diabetes patients are highly motivated to find easy cures for their condition.
• Lack of Digital Literacy: Vulnerable consumers fail to notice poor lip-syncing and red flags.

Process and Methods

• Reactive Moderation Process: Platforms rely on manual reporting instead of proactive deepfake prevention.
• Whack-A-Mole Takedown Loop: Deleted scam videos are immediately re-uploaded under different accounts.

Regulatory Environment

• Weak Social Media Accountability: Lack of strict regulations holding platforms liable for hosting scam ads.
• Inadequate Consumer Protection: Current laws fail to deter cross-border AI-generated financial scams.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide consistent accounts of the deepfake scam, identifying specific victims, financial losses, and the nature of the deceptive content. The role of AI in generating the video is clear, and the primary domain classification of fraud/scams is highly unambiguous.
• Ambiguities identified: The exact AI model or software used to generate the deepfake is not specified.
• Alternative interpretations: None. The incident is clearly a malicious deepfake scam targeting vulnerable individuals for financial gain.

An AI-generated deepfake scam in New Zealand impersonated a prominent journalist and medical expert to promote a fraudulent diabetes cure. While posing a direct health risk to individuals who stopped medication and causing minor financial losses, the incident represents a minor national security threat primarily concerning the erosion of societal trust in media and health institutions.

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

Threat characteristics

• Imminence: Long-term. Represents an ongoing strategic concern regarding the proliferation of synthetic media scams rather than an imminent national security crisis.
• Autonomy: Human-controlled. The AI tool was used by human operators to generate content, with no autonomous decision-making or execution by the system itself.
• Novelty: Evolved capability. While deepfake scams are an established threat, the level of sophistication and specific targeting of local trusted medical experts represents an evolved capability.

Impact by dimension

• Physical security: Negligible. No threat to critical infrastructure, physical systems, or national-scale human safety. The health risks are localized to individual scam victims.
• Information security: Minor. Involves sophisticated deepfakes of public figures, but the intent is financial fraud rather than state-sponsored information warfare or intelligence compromise.
• Sovereignty: Negligible. No impact on sovereign authority, government decision-making, border control, or electoral systems.
• Economic security: Negligible. Minor financial losses to individuals with no systemic impact on strategic industries, financial systems, or technological advantage.
• Societal stability: Minor. Erodes public trust in media and health institutions by using synthetic media to impersonate trusted experts, though limited to a minor scale.