AI Safer from Hackers

(See details at the link)
A new way to train AI systems could keep them safer from hackers
Technology Review: Blogs: Mims's Bitsby Karen Hao
Artificial intelligence
" ... The research: Bo Li (named one of this year’s MIT Technology Review Innovators Under 35) and her colleagues at the University of Illinois at Urbana-Champaign are now proposing a new method for training such deep-learning systems https://arxiv.org/pdf/2002.11821.pdf to be more failproof and thus trustworthy in safety-critical scenarios. They pit the neural network responsible for image reconstruction against another neural network responsible for generating adversarial examples, in a style similar to GAN algorithms. Through iterative rounds, the adversarial network attempts to fool the reconstruction network into producing things that aren’t part of the ground truth, and the reconstruction network continuously tweaks itself to avoid being fooled, making it safer to deploy in the real world.  ... "\