Ajit Jaokar: Artificial Intelligence #6: Security threats to artificial intelligence models and how to mitigate them
2/06/2021 – 08:29
Welcome to edition #6 of the newsletter
We are nearing 10,000 subscribers. Thanks for your support as ever
This week, we announced a new course Virtual Reality and Augmented Reality applications which covers the development of professional VR and AR applications – combining with digital twin. This is based primarily on the Unity platform.
I will discuss more on the technologies behind this course soon. Its also the first time we are working with the Oxford Robotics Institute, a part of Oxford that does some pretty cool work on autonomous robotics see this introductory video on Oxford Robotics
If you are interested in these areas (VR, AR, Digital Twin etc) pls have a look or message me
Today, we will cover security threats to artificial intelligence models and how to mitigate them. I often get good insights from speaking to my students in Oxford. Last week, François Ortolan, from our course recommended me a paper about protecting Artificial Intelligence models against threats and attacks.
This is an important topic, both for our teaching and also for some of the clients I have worked with. Protecting AI models from cyber threats is not fully understood by most people – but it’s getting increasingly important.