Umjetna inteligencija uči prepoznavati hakerske na

VoodooShield je vec ubacio nesto slicno u zadnjoj beta verziji.

Evo sta kaze developer VoodooShield-a:

What VoodooAi does it a lot more similar to the way Cylance utilizes Ai... the main differences being that they use Amazon Web Services for their modeling, whereas VoodooAi uses IBM Watson in the initial stages, and Microsoft Azure for the final analysis. Well, the other differences are they have teams of data scientists (as opposed to just Molly and I), and their product is only available to the enterprise. I would be curious to compare the two head to head... I have only been able to make 3 comparisons, and VoodooAi did very, very well. I think Ai is going to play a huge role in computer security over the next few decades, but it is not the silver bullet... it is never going to be absolutely perfect, but it is quite adept in finding the really bad unknown and zero day malware. So when combined with other technologies, it is extremely useful.

Zasad samo to, mozda se jos negdje spominjalo o tome na tom forumu (Wilderssecurity) odakle sam C/P ovo gore ali mi se neda trazit.

Eventualno bi forumas @siketa mogo znat jos malo o tome.

Mozda jos ovo (ako sta pomaze):

What MIT is doing is quite similar to to what Palo Alto Networks (among others) is doing with Ai... it is more on the network level as opposed to endpoint protection like VS / VoodooAi. I am a little surprised that they are so far only detecting 85% of malware... although that is certain to get a lot better, I am sure they have some of the best data scientists in the world that will soon achieve a 99% or so detection rate.