Has anyone attempted to create a worm virus that when it self replicates it intentionally creates randomized changes into the code of its offspring to mimic the random mutations of biology that leads to evolution?
>>53556438
God did.
>>53556438
>Has anyone attempted to create a worm virus that when it self replicates it intentionally creates randomized changes into the code of its offspring to mimic the random mutations of biology that leads to evolution?
Yes
All advanced modern viruses and worms do this in order to avoid detection from anti-virus programs.
>>53557015
I think op is interested in biological type evolution where features of the program are added or removed. Like a real mutation. I too am interested in this
Very interesting thread, bumpity
>>53557312
>features of the program
That's not real mutation though. Real mutation are just random changes to source/machine code/DNA which might result in "traits" that renders the program useless or actually does something useful.
However, you need some sort of non-random selection that deems what traits are desirable and what are not in order for it to be evolution.
>>53556438
Not enough research was put into it.
Do you even have the slightest clue on how would it work ?
>>53557660
>>However, you need some sort of non-random selection that deems what traits are desirable and what are not in order for it to be evolution.
Evolution doesn't care for desirability, it can easily cripple a species as much as it can help, in the long term.
I wanna make a computer virus that infects humans too.
>>53557891
Exactly, the environment is what selects which traits are successful
>>53556438
dear diary, the ass was fat
>>53557948
But then you need this sort of environment, otherwise you just end up with a self-replicating program that adds in random bugs causing it to crash.
This wouldn't be hard with some languages that have a programmable assembler. Off the top of my head I know that this could be possible in Java and either .NET languages. You can literally recompile the program itself from within itself changing everything, and you wouldn't be able to tell the difference.
Metaprogramming is a beautiful thing
>>53556438
Yes, your grandma.
>>53557312
OP here, in biology mutations are completely random and most of them are harmful or benign but a small percentage of them become beneficial and lead to the evolution of the species.
>>53557660
>However, you need some sort of non-random selection that deems what traits are desirable and what are not in order for it to be evolution.
Well in this situation it would be the continued survival of the virus. If you create a virtual sandbox for a program to live in then you would definitely have to.
>>53558017
Which is in a way its environment.
if it generates a bug and crashes then it was not fit to survive in its "digital" environment.
>>53558555
MY TRIPS DONT LIE
i thought about this for awhile, considering the possibility of a rogue dangerous AI evolving from it. It's theoretically possible, however realistically not.
Issues:
computational space and time.
bugs
self destruction of environment/self
Basically you'd have to create an environment with millions of little self replicating, self modifying programs.
Then maybe have some sort of natural selection mechanism. and maybe give them knowledge of the grammar of the language they're written in, and protect the core of the program from being modified.
