[dox20]

I'd like to know what you all think about web vulnerabilities and whether they will gradually disappear due to AI-based protections. Will it be a problem? or will ai's get bypassed? tell me about it

[sabat]

I'd like to know what you all think about web vulnerabilities and whether they will gradually disappear due to AI-based protections. Will it be a problem? or will ai's get bypassed? tell me about it

AI isn't actually artificial intelligence because it just consists of huge databases full with things humans made (including human errors)

so if you feed an ai with a dictionary you might think that it wouldn't be able to do grammar mistakes

but if the human who made the dictionary wrote the word "coffee" with only one "e" instead of two the ai will output the word just like it was given to it and don't think its wrong

then a different person could come and see that the word "coffe" is written wrong and exploit the sentence the ai made (this is just a example no one is going to exploit a sentence)

but the same thing can happen with actual code


now you could say

"but code library are code not just sentences, every word has a function. If there was a grammar mistake it would immediately get noticed"

response to that is that ofc grammar mistakes should get noticed faster but in the topic of code there are a different kind of mistakes people make besides grammar mistakes

and if its so clear and errors would get detected asap why are code library still being updated

[dox20]

I'd like to know what you all think about web vulnerabilities and whether they will gradually disappear due to AI-based protections. Will it be a problem? or will ai's get bypassed? tell me about it

AI isn't actually artificial intelligence because it just consists of huge databases full with things humans made (including human errors)

so if you feed an ai with a dictionary you might think that it wouldn't be able to do grammar mistakes

but if the human who made the dictionary wrote the word "coffee" with only one "e" instead of two the ai will output the word just like it was given to it and don't think its wrong

then a different person could come and see that the word "coffe" is written wrong and exploit the sentence the ai made (this is just a example no one is going to exploit a sentence)

but the same thing can happen with actual code


now you could say

"but code library are code not just sentences, every word has a function. If there was a grammar mistake it would immediately get noticed"

response to that is that ofc grammar mistakes should get noticed faster but in the topic of code there are a different kind of mistakes people make besides grammar mistakes

and if its so clear and errors would get detected asap why are code library still being updated

ohh you have a point

[ShikataGanai]

I see that only idiots will be replaced by AI cause the AI itself doesn't know how to think independently, so it's not gonna fix anything the human can't fix it's all about feeding him a huge data.

[DredgenSun]

Nah, they'll get worse I think. Wouldn't suprise me that companies create vulnerabilities to sell solutions,

[reflex]

Just like they’ve changed many things, artificial intelligences will also transform hacking and cybersecurity. First of all, AIs are trainable machines, and if you train them for cybersecurity, they can perform defense operations. But here’s the issue: an AI that can defend can also attack.
Keep that in mind. With the widespread use of AI, cyberattacks have become more complex, and much more effective malware has emerged. Considering all this, AI will partially replace cybersecurity experts and hackers, significantly changing the industry and making it even more challenging.
Still, don’t forget—AI is like a baby right now. It’s only just starting to crawl.

This forum account is currently banned. Ban Length: Permanent (N/A Remaining)
Ban Reason: Dishes out second hand retardation | http://c66go4clkqodr7tdjfu76jztjs7w7d3fajdeypxn73v4ju3dt7g5yyyd.onion/Forum-Ban-Appeals if you wish to be unbanned in the future.

[dox20]

Just like they’ve changed many things, artificial intelligences will also transform hacking and cybersecurity. First of all, AIs are trainable machines, and if you train them for cybersecurity, they can perform defense operations. But here’s the issue: an AI that can defend can also attack.
Keep that in mind. With the widespread use of AI, cyberattacks have become more complex, and much more effective malware has emerged. Considering all this, AI will partially replace cybersecurity experts and hackers, significantly changing the industry and making it even more challenging.
Still, don’t forget—AI is like a baby right now. It’s only just starting to crawl.

you are right ai has a long way to go rn

[ElPeponcio]

Look, not all pages uses AI or will use it, since they think that they won't be hacked in a long time. Most of web admins forget to fix some thinks because they watch tutorials on youtube and etc and those little thinks can fuck up their web pages

[Blinkers]

This is never gonna happen. Even A.I can't make a code 100% secure. Just like a thief finds a way to open a locker without any tool.

[sunfryre]

hmm not really. i actually think that because of AI and how easy it is to get access to tools, more people will get into hacking and increase targeting