Technology is constantly, deepfakes are at an extraordinary stage of development in recent months, according to one expert.
It predicts that in the next six to twelve months we will have our faces manipulated extremely realistically in videos and photos.
Deepfakes have been quite popular in recent times due to the numerous apps produced for the purpose. In addition, developments in this technology have been noticeable and the results achieved often captivate users. In short, a deepfake is a fake representation of a person’s face in a video or photo.
Now one expert in this area says he is very excited about the near future. According to their predictions, shared with CNBC, in the next six to twelve months we will have perfect deepfake whose handling will be completely unnoticed. In addition, it is estimated that deepfake creation tools are available to all users with simple steps.
It’s still very easy to see most deepfakes with the naked eye. But there are also examples that are really convincing. It will soon reach the point where it cannot be detected, so we need to look for other types of solutions. In a way, we already know how to do it, it’s just a matter of training with more data and implementing it.
For the deepfake expert, application development such as ZAO is strongly driving the development of this area. Thus, the Machine Learning used for these skills is reaching levels never seen before.
Hao Li recently held a demonstration of the current state of technology. At an MIT conference last week, the researcher showed a video of himself in which he implemented a deepfake by Vladimir Putin. The reason for doing this was to show the evolution that facial manipulation has undergone in recent times.