ML Security: Why should you care? // Sahbi Chaieb // MLOps Coffee Sessions #51
MLOps.community
00:00
Machine Learning Security - Why Should We Care?
i love thata journey. And maybe we can talk now about some of these reasons why we should even care about keeping our models secure? Yeso, when we're talking about machine learning security, first we have to be aware that machine learning systems are actually softer pieces. So as soon as your model is subject to attacks, it's actually with the data that is used, which is a threat as well. We know that this service could be rel problematic interms of privacy, first and formusts. It d be another hypo problem, affecting the model by making it generate false predictions. For example, its alesan social mediate ise, a fit news detector. S
Play episode from 09:08
Transcript
