Deep…what?
Deepfakes use advanced artificial intelligence and machine learning to create hyper-realistic audio and video footage that makes people appear to say or do things that never happened. This technology has the potential to distort perceptions of reality, spread misinformation and undermine trust in digital content.
A deepfake showing a young wounded Israeli woman clinging desperately to the arms of a soldier. This impressive image captures the feelings of a war-torn region extremely well. But it is an image generated by artificial intelligence
This deepfake recently caused a stir in the media. It shows a picture of a girl from Ukraine standing in front of the ruins of a city with her cuddly toy in her arms. This image is also AI-generated and was never photographed in this way.
This deepfake was distributed on the Internet without indicating that it was an artificially generated image. It shows an alleged explosion in Gaza
Developments in the field of artificial intelligence (AI) have seen a remarkable acceleration in recent years, particularly in the area of deepfakes and generative AI. This rapid development is mainly due to advances in the areas of machine learning, computing capacity and data availability.
Review:
Development in recent years
Beginnings of the deepfakes
The first known deepfakes appeared around 2017. At that time, the creation and editing of deepfakes still required considerable technical knowledge and computing resources. These videos were often of low quality and easy to recognize as fakes.
Qualitative leaps
Within a few years, algorithms and software for deepfakes have made enormous qualitative leaps. Improved algorithms, especially those based on GANs (Generative Adversarial Networks), made deepfakes more realistic and more difficult to identify. The release of user-friendly tools and apps made the creation of deepfakes accessible to non-experts.
Commercialization and distribution
Recent years have seen an increased commercialization and proliferation of deepfake technology. Platforms offering real-time deepfakes for video conferencing, entertainment and even educational purposes have become popular. At the same time, deepfake detection technologies improved, leading to a race between the creation and detection of deepfakes.
The development of AI and deepfakes is expected to continue to progress rapidly.
Exposing deepfakes:
our mission
In the face of this increasing threat, we have made it our mission to contribute to the defense against these dangers. With our passion for technology and an understanding of the mechanisms of artificial intelligence, we have initiated the development of a tool specifically designed to reliably detect and identify deepfakes. This project is not just a technical endeavor, but a necessity to maintain integrity and trust in digital communication.
To achieve this ambitious goal, we have realized that close collaboration between industry and academic research is essential. Innovations in AI technology and its applications are not created in a vacuum – they require expertise, resources and a network of committed partners who are willing to invest in the future of our digital world.
For this reason, we decided to take part in the Founders Foundation in Bielefeld. This initiative provides a platform for founders to not only gain the necessary business knowledge and strategic insights, but also to network with like-minded people, investors and subject matter experts. Through these gatherings, we hope to find the right partners – those who share the vision of tackling the challenges posed by deepfakes and joining forces to foster a safe and authentic digital environment.
