Month: September 2022

CHIMERIC CODE. "In the event of coinfection, the excised viral genome might be a combination of more than one viral strain leading to the creation of chimeric viral particles." https://lnkd.in/dGC6A7c View in LinkedIn

DORMANT CODE. "During a mode of viral reproduction termed “lysogenic”, viruses such as bacteriophages integrate their genome into the host’s genome and become dormant. When such viruses convert back to the lytic (aggressive) mode of reproduction, they excise themselves from the genome." https://lnkd.in/dGC6A7c View in LinkedIn

GIANT CODE. "Imagine a random change that results in the macro code becoming very large, either as the result of several layers of merged code or additional polymorphisms. This can easily exhaust typical scanning engines for macro viruses due to memory limitations reserved for the macro code within the engines. As a consequence, an evolved copy can potentially escape just by being too large." https://lnkd.in/dGC6A7c View in LinkedIn

IS SMALL SUFFICIENT? "The question that arises from these examples is whether small malware code changes—even if collected under a single malware implementation—are sufficient for the evolution of the functional complexity exhibited by biological viruses." https://lnkd.in/dGC6A7c View in LinkedIn

EMERGENT PROPERTIES. "We can imagine the emergence of arbitrary complexity in autonomously evolving computer viruses." https://lnkd.in/dGC6A7c View in LinkedIn

THE LEAPING ACCUMULATION OF SMALLS. "The myriad examples...of unintended virus behaviors due to random changes are all examples of more complex behaviors originating from simpler ones through small programming jumps." https://lnkd.in/dGC6A7c View in LinkedIn

LARGE CODE SOURCE. "As in biology, the collection of software and APIs in an operating system environment is sufficient for providing a pool of features that a virus can choose to sample." https://lnkd.in/dGC6A7c View in LinkedIn

CODE NATURAL SELECTION. "Those features that provide the virus with an evolutionary advantage (e.g., escaping detection, infecting different platforms, etc.) will be propagated, leading to unanticipated complex behaviors." The survival of the fittest code, in Darwinian terms. https://lnkd.in/dGC6A7c View in LinkedIn

"COMPLEX FUNCTIONALITY can evolve from much simpler functions in Artificial Life platforms, over the course of a few days, and complex adaptations have been seen to emerge after tens to tens of thousands of generations, easily within range of weeks of malware evolution." Aeons of evolutionary time which are necessary in biological systems are not required in digital ecosystems. https://lnkd.in/dGC6A7c View in LinkedIn

PREREQUISITES OF COMPLEXITY IN ARTIFICIAL LIFE. "First and foremost are of course the three conditions for Darwinian evolution themselves: replication, variation and selection. Of those, replication and selection are naturally present, and we argued above that variation can occur via a variety of mechanisms. However, in order for sufficient variation to be present so that selection can act efficiently, population sizes should be large." https://lnkd.in/dGC6A7c View in LinkedIn