Playing defence is hard
If you know in advance what tools your opponent is using, you can beat them.
Colour me astonished.
Adversarial models, already known to defeat the artificial intelligence behind image classifiers and computer audio, are also good at defeating malware detection. Last year, researchers from NVIDIA, Booz Allen Hamilton, and the University of Maryland probably felt justifiably pleased with themselves when they trained a neural …
When I submit the binaries of my own code to Virus Total, guess who's crap has the highest false positives? The AI-AV's. And guess which companies either have no false positive reporting mechanism, or otherwise seem to be the most unresponsive if they do have a false positive reporting mechanism? That's right, the AI-AV companies.
The One Time Pad 0Day Vulnerability Exploit Rules .... with Novel Intelligence and True Views Autonomously Trumping Fake and Terrible News.
What possible ineffective and self-defeating defence is there other than a dogged ignorant and arrogant refusal to acknowledge changed and rapidly moving virtual reality targets creating and/or presenting future events which lead to reactions which do vain battle against HyperRadioProACTive IT.
"neural networks are subject to adversarial attacks."
Such as the Berryman Logical Image Technique