Deepfake Medical Scam Targets Ophthalmologist on Facebook, Highlights AI Ad Fraud
An Auckland ophthalmologist has become the victim of a deepfake video promoting a fake medical product on Facebook, created using generative AI to manipulate her image and voice. The incident highlights challenges for platforms like Meta in moderating sophisticated AI-generated scams, particularly those using "dark ads" to target vulnerable users. The physician has experienced frustration with Meta's automated reporting tools, which are not equipped for advanced deepfakes, and is advocating for better collective protection against such scams.
Key Takeaways
- Dr. Divya Perumal's identity was deepfaked to endorse an unregulated 'sight-saving' product on Facebook, claiming it cured conditions like glaucoma and cataracts.
- The deepfakes likely used source material from Re:Vision's 117 educational YouTube videos to recreate her voice and image.
- Scam advertisements were distributed as 'dark ads' on Facebook, making them invisible on public pages and difficult for Dr. Perumal to detect.
- Meta's automated reporting tools were inadequate for the sophisticated deepfake, designed for standard copyright issues rather than AI-generated identity theft.
- Ad fraud, particularly involving deepfakes and 'dark ads,' costs advertisers tens of billions annually, with Juniper Research projecting losses over US$170 billion by 2028.
Why It Matters
This incident highlights the growing threat of sophisticated AI-driven deepfake campaigns for ad fraud and misinformation, particularly in vulnerable sectors like healthcare. Platforms like Meta face increasing pressure as their existing content moderation and reporting systems prove ineffective against advanced generative AI. The use of 'dark ads' complicates detection, forcing medical professionals onto the front lines of defense. The industry should watch for collaborative efforts between platforms and professional bodies to develop more robust detection, reporting, and enforcement mechanisms for AI-generated scams, particularly as generative AI tools become more accessible.
Read full article at nzoptics.co.nz
Enjoy our coverage?
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source