YouTube Removes Over 1,000 Deepfake, AI-Generated Scam Ad Videos
Amid the rising concerns over the misuse of artificial intelligence (AI) by scammers, YouTube has deleted more than 1,000 deepfake video ads from its platform.
According to reports, these deepfake videos include many of them that had millions of views on YouTube.
YouTube's deepfake ads
These deepfake video ads created using AI featured famous personalities like Taylor Swift, Steve Harvey, and Joe Rogan promoting Medicare scams on YouTube.
The removal of the deepfake videos came after an investigation by 404 Media, into the extent of such misleading content on YouTube.
YouTube said that it was fully aware of the issue and is actively working to prevent the proliferation of celebrity deepfakes.
The Google-owned platform said it has been investing resources to combat these AI-driven celebrity scam ads.
Deepfakes have become a major cybercrime in recent times, with celebrities and common people equally falling victim to it.
According to the latest research from cybersecurity firm Deeptrace, about 96 per cent of deepfakes are pornographic, and they almost always portray women.
Taylor Swift's deepfake image goes viral?
Another report by 404 Media also looked into an explicit deepfake of Taylor Swift that went viral on X.
One post of the deepfake porn of Taylor Swift had more than 45 million views and 24,000 reposts before it was removed from X.
According to 404 Media, the images of Taylor Swift originated in a Telegram group dedicated to sharing "abusive images of women and later reached X.
Who Taylor Swift deepfake was created?
What makes the images an even bigger concern is that, the report said they were not created by training an AI model, but by hacking tools like Microsoft's Designer to override safeguards designed to stop tools from generating images of celebrities.
"Members of the group shared strategies for subverting these safeguards by avoiding prompts using "Taylor Swift" and instead using keywords like "Taylor 'singer' Swift." They were then able to generate sexualized images by using keywords describing "objects, colors, and compositions that clearly look like sexual acts," rather than attempting to use sexual terms," the report said.
AI the new tool of cybercriminals
With free AI tools that can generate realistic-looking deepfake videos and images becoming widely available, AI is turning out to be a goldmine for online scamers.?
Tech giants including Google, Facebook and Microsoft have said that they are improving their capabilities to detect deepfakes and AI-generated content on thier platforms.
For more news and current affairs from around the world, please visit?Indiatimes News.
