PhD vacancy at MultiX


Do you want to do research at the frontier of Artificial Intelligence to support forensic investigators? The MultiX group of the Informatics Institute of the University of Amsterdam in collaboration with the Netherlands Forensic Institute is seeking a PhD student to develop forensically sound methods to determine whether a video is fake or genuine. With the advances in AI and in particular Generative Adversarial Networks, the quality of deep fakes is improving so rapidly that the differences between fakes and real videos are subtle. We aim to develop detection techniques that stay ahead of what is possible in generating deep fakes, leveraging the best in computer vision as well as the knowledge of forensic experts who interactively can steer the analysis process through appropriate visualizations. See for more information the MultiX group and the Netherlands Forensic Institute.

What are you going to do?

You are expected to:

Perform research in the detection of deep fakes, publish and present the results in international journals and leading conferences, and to write a PhD thesis;

  • develop tools and to evaluate them with expert users;
  • collaborate with other researchers in our Multimedia Analytics (MultiX) group;
  • work one day a week (when this is possible again) at the Netherlands Forensic Institute to collaborate with forensic experts; contribute to our education activities as assistant to a course and/or guiding students in their thesis work.

What do we require?

  • A master’s degree in Artificial Intelligence, Computer Science or related field;
  • a strong scientific expertise in Machine Learning and Computer Vision;
  • demonstrated experience in using and developing deep learning based methods in python;
  • the ability and willingness to acquire expertise on forensic processes;
  • good writing and communication skills in English;
  • impeccable behaviour: part of the hiring procedure is a security screening.

This position is also available on the official university website: