The knowledge gained from the pre-lab and hands-on lab in this module allows students to further explore the threats available through deepfake phishing and their solutions.
Using VAEs to simulate Deepfake Phishing:
Recall in the pre-lab for this module, we reviewed Variational Autoencoders. Despite their usefulness in anomaly-detection, Generative AI could use them for deepfake phishing. For this post-lab, you are to use VAEs to generate phishing emails that can be distributed through email servers, making them appear as routine traffic. Additionally, combine the use of Natural Language Processing (NLP) and anomaly detection to mitigate the problems with VAEs. Review the information found in the pre-lab for this module for further information. You may use Google Colab to implement this.