On August 22, 2023, STEKOM University held an international seminar titled "The Incredible Creativity of Deepfakes and the Worrying Future of AI." The seminar discussed deepfake technology and its potential impact in the future. The event was hosted by Novita, Vice Rector for International Relations at STEKOM University.

The seminar was attended by a number of international speakers who are competent in the fields of technology, cybersecurity, and business. In her presentation, Migunani M.Kom, Head of the Information Systems Department at STEKOM University, went in-depth about deepfakes and their potential impact. Here is a summary of his presentation:

Introduction to Deepfake
Migunani started her presentation by explaining the basic concept of deepfake. Deepfake is a software application that utilizes machine learning algorithms to create fake videos that are continuously improved by mimicking individual expressions and voice modulation. This makes deepfakes increasingly realistic and difficult to distinguish from the original video.

What is Deepfake?
Migunani describes deepfake as a hyperrealistic synthetic medium where the face of an individual in a photo or video is replaced by the face of another person. There are two types of deepfakes: face-swapping and puppet master (character creation with total control).

Deepfake Examples
Migunani gave some famous examples of deepfakes, including a video of Elon Musk being turned into Tony Stark. He also shows how deepfakes can be used to replace someone's face in an image or video with someone else's, creating a very convincing fake video.

How to Make a Deepfake
Migunani explained the process of creating a deepfake using machine learning techniques and GAN (Generative Adversarial Network). The process involves compression and face replacement, and in the end, a deepfake is created.

Potential Uses of Deepfakes
Deepfakes are not just about crime. Migunani outlines some positive potential uses, such as the movie and advertising industries, where deepfakes can be used to replace actors or create more realistic multilingual ads. It can also create interactive movies where the audience can choose the players in the story.

Negative Impacts and Regulation
Migunani outlined some of the potential negative impacts of deepfakes, including the potential spread of false information, election manipulation, and fraud. He also discussed the role of regulation in controlling this technology and how governments and agencies are considering measures to address deepfake risks.

The Troubling Future of AI
Migunani ended his presentation by discussing the worrying future of artificial intelligence (AI). He noted that the concern is not only about artificial intelligence becoming smarter than humans, but also about its unpredictability. A smart or dumb AI can quickly become a threat, and there is uncertainty about how to control it. Migunani also highlighted the potential for AI abuse and the role of regulation in managing this technological development.

The seminar revealed how important a deep understanding of deepfakes and their future impact is. Thoughtful regulation and regulation can help address the risks associated with this technology. It is our hope that advancements in artificial intelligence can positively benefit the world community without compromising the security and veracity of information.