Reflection of Deep Fakes

From watching the presentations, one topic that specifically stood out to me was deep fakes. According to Shruti Agarwal and Penny Ferret from the University of Berkeley, deep fakes are the manipulation of digital media using digital effects to create AI-synthesized content. During the presentation, it was also discussed that there are three types of deep fakes: Face Swap, Lip Sync, and Puppet Master deep faking. Personally, I was unaware of the widespread use of deep fakes before recently, and I was surprised to learn that this technology was used in the Star Wars movie to digitally recreate Princess Leia, originally played by the late Carrie Fisher.

Another example that resonated with me was the deep fake of Joe Kwon Oliver, a victim of the Parkland shooting in 2018. His parents, using AI, were able to create a video that replicated his voice and likeness to promote gun safety. What stood out to me was the emotional impact. Even though it was a digital recreation, the fact that his parents brought him back to deliver a message so personal to them made it incredibly powerful. Seeing him "speak" again gave the message a weight that wouldn’t have been as impactful if it came from a typical public service announcement.

While it’s fascinating that advancements in AI are leading to more realistic deep fakes and leading to new forms of creativity, it also raises serious concerns. As these AI-generated videos become more lifelike, the line between what’s real and what’s fake begins to blur, making it easier for scammers and sick individuals to exploit the technology. From further research, I found a report on Stanford’s website that warns individuals to “be wary of phone calls or videos that appear to come from trusted colleagues or senior executives when they involve unexpected demands or requests for financial transactions.” This highlights the darker side of deep fakes and their potential for fraud and manipulation.

In the presentation, the speaker mentioned how Penny Ferret found that, in 2022, there were over 1,000 LinkedIn accounts with profile pictures that were synthetically generated. These accounts would message users, trying to promote products or services, all while hiding behind AI-generated faces. This really stood out to me because I’ve seen it happen on my own LinkedIn—accounts with sketchy profiles reaching out with offers that just didn’t feel right.

It is clear that deep fake technology isn’t just about videos; it’s also being used to create fake identities on social media. The mix of AI and social engineering makes it hard to know who or what to trust online. As these fake profiles get more convincing, it becomes easier for people to get tricked into scams or misinformation.