How biometrics and liveness detection are evolving to protect us from deepfakes and chat scams
For years, deepfakes were considered a looming fraud risk, but the launch of ChatGPT shows that we will soon be able to interact online with machines that are “almost” indistinguishable from humans. Chatbots will be domain experts on call, providing technical support and business advice; they will collaborate on projects and take action on Zoom calls. The deep neural networks behind the deepfakes will be used to give a human face to our virtual adversaries.
Unfortunately, these methods will also be used to attempt “deep fraud” and commit crimes. But liveness detection has come a long way since it was used to detect rubber thumbs. In the brave new world of human-machine interaction, biometrics and liveliness are critical to protecting our interactions with machines from intruders.
– New use cases for human-computer interaction based on deep learning and generative AI – How to use biometrics to mitigate security threats from hyper-realistic robots and fakes – How to use liveliness detection to prevent presentation attacks and intrusion attacks – Human and machine fluency impact on security and user experience
Post time: Feb-15-2023