Call us now:
A great, and simple, explanation of why AI and modelling is difficult.
Extrapolate this to cyber. where the data sets are huge, where the relationships between data points withing security domains are difficult to correlate. For example, an unpatched vulnerability does not necessarily mean you will get hacked and more over what the hack will result in. The probability models need to learn about interactions and Monticarlo simulation has its limitations. (How do you correlate the multiple threat vectors, security domains, access management, V&P, obsolescence and people and an outcome of being hacked, to many inputs).
It’s a reason why we are a long way off any meaningful models which will predict a hack. But we are getting closer to modelling the impact through residual risk modelling.
View Original Post
https://www.linkedin.com/feed/update/urn%3Ali%3Ashare%3A6568569719153049600