Sign in to follow this  
Followers 0
mrfishopolis

AI and Hacking project/site

22 posts in this topic

I would like to start a site dedicated to the use of AI (Agents, Algorithms, and Techniques) that could be used in a security related content. This is just in the conceptual phase. I a few topics I have done some initial research into, but I would like the community input on topics and possible directions of this project, also would this be of interest.

Here is the basic thesis of the site:

Explanation and application of traditional AI solutions to Security and Hacking problems.

Just a Sample of the topics that I have been thinking about.

*How to use all that data you've minned, AI based filtering.

*Rule based AI and nmap.

*Digital Swarms as a thinking army of Attack agents.

I hope to provide not only this but quick and dirty introductions to AI topics that will be used and explain on a casual level what is happening, and provide code and diagrams where necessary.

Let me know what everyone thinks.

0

Share this post


Link to post
Share on other sites

I would be definitely interested in helping yuo get something running. Drop me an email with a plan at s0u1d13r AT gmail DOT com

S

0

Share this post


Link to post
Share on other sites

i am also interested. please email verbal at gmail. thanks.

0

Share this post


Link to post
Share on other sites

Let's get something started. I've got a few AI e-books. If we come up with a few specific goals it should be manageable.

PM me or email me through the board.

0

Share this post


Link to post
Share on other sites

Sounds like a good idea. I'll help.

Edited by matt
0

Share this post


Link to post
Share on other sites

Why is everyone throwing out their email address? Discussion is supposed to happen on here, it's a public forum. Don't you just hate searching for something only to find that people were actually just asking a question about it and then everyone who replies asks someone to email it to them once they find it?

0

Share this post


Link to post
Share on other sites

I'd be intested about learning AI, I really don't know much. But that'd be really awesome.

I was thinking about getting an AI book, but all of them are around $50-60.

And how would we use AI for security? It sounds like a cool idea, but it also sounds like Neuromancer.

It'd just be cool to do AI, AI in security would make people think that "I, Robot" is going to happen.

I wonder how many science fiction references I can make in one post...

0

Share this post


Link to post
Share on other sites

To everyone who replied I will be setting a up a list and I will send updates to those intrested once I get to a point where I figure out just what I am going to need for help and what skill sets I will need. If there are any more intrested private message to my account, I will try to send out at least a introductory message at the end of the week.

To give you a quick tutorial for those new to AI and the techniques, not all AI is super high level stuff, it is for the most part just a subset of algorithms that are used for things as simple as finding the shortest path on a map/graph, to slightly more complex things making decisions on information provided.

I want to make accessible articles and tutorials that provide examples or solutions that intresting to hackers, really it is just another way to broach this subject, becuase most computer science articles involves examples consisting of salesman, foo/bar, animals and the like. So to put it simply Providing Articles on topics that provide practicle examples, and if not practicle then interesting ideas.

Edited by mrfishopolis
0

Share this post


Link to post
Share on other sites

I have a few books on AI. Add me to your list I'm interrested in what you have in mind.

dexter[at]synthetica.org

But I do agree that most of the discussion should take place in an open forum.

0

Share this post


Link to post
Share on other sites

ok, i guess i should start the ball rolling in the open forum. just some random thoughts...

when you say use AI/ML for security, the first two things that come to mind is defense and offense. for example, in the category of defense, it is fairly straight forward to conceive of a mechanism that implements an IDS at the OS or network level. to do this, you would basically train your knowledge base with "acceptable" behavior. this includes anything from network flows, syscalls, etc. then if someone is doing something "unacceptable", your knowledge base will flag it. depending on what AI/ML algorithms you use, you will either give it human positive/negative feedback or ... you wont.

offense is a little tricky. one place i think some AI/ML will be beneficial is worms and viruses. you can use information learned in the wild to make better decisions about which boxen to spread to or which exploits to use. the problem with this is that each spawning will need to carry around with it the information that its learned. from what i know, the "brain" isnt going to be small and that is going to be a problem.

lastly, for something completely different, ML and data mining is just a good way to get information from large data sets. we may be able to find innovative ways to use this information to make better decisions about attacks. maybe we can even find vulnerabilities this way ;).

ok, that's all i got right now. hopefully it'll spark some discussions.

as for information about AI/ML, i have a bunch of textbooks and papers. once a website gets going (maybe a wiki), i can write an outline of AI/ML topics and fill out most of it in. i'm sure other people can help =).

0

Share this post


Link to post
Share on other sites

Well I got a basic front page up, hopefully I will still get that email out to everyone intrested. For now if binrev doesn't mind I will be posting updates and discussion in this post, till I get a forum up.

http://www.mybrainisgone.com

0

Share this post


Link to post
Share on other sites

The whole project sounds fascinating. I can't claim to be any use, but I'll be watching with interest and hopefully learn lots.

0

Share this post


Link to post
Share on other sites

I too am very interested, and might even contribute with some code myself, once someone gets some decent tutorials out there. Please add me to the mailing list

thegreatmausa[at]gmail[dot]com

0

Share this post


Link to post
Share on other sites
Well I got a basic front page up, hopefully I will still get that email out to everyone intrested.  For now if binrev doesn't mind I will be posting updates and discussion in this post, till I get a forum up.

http://www.mybrainisgone.com

i dont know if you care, just wanted to let you know that your page doesnt render correctly in safari (Version 2.0 412.2). i'm running Tiger 10.4.2. the table with all the content stretches all the way to the right side of the screen.

0

Share this post


Link to post
Share on other sites

Is anything going to happen with this?

0

Share this post


Link to post
Share on other sites

Sorry I have been dead last couple weeks Trying to get my post college life in order. I should be getting some info out this week. I have the base plan 95% complete.

Edited by mrfishopolis
0

Share this post


Link to post
Share on other sites
Sorry I have been dead last couple weeks Trying to get my post college life in order.  I should be getting some info out this week.  I have the base plan 95% complete.

its cool. keep me in the loop. come hang out at #binrev sometime so we can chat and you can get your hazing. =)

0

Share this post


Link to post
Share on other sites

Awesome, because I don't know much about AI, and I want to learn stuff. The books that I see at Barnes & Noble are always too expensive, so I only get to read them briefly when I go there.

0

Share this post


Link to post
Share on other sites

Sorry I don't know if this is dead or not, so I'll bump it to see :X.

I'd love to help out, I've been playing with the idea for a little bit too.

I'm not good at AI, but I have created a small nueral network for linearly separable problems, used genetic algorithms to get a best guess for the NP-hard TSP, and currently I'm using data mining and rule induction algorithms using stock quotes as my data.

My idea doesn't really use AI techniques, it's related to genetic algorithms in that it's biologically based. I think it would be cool to make a worm that constantly modifies itself and tries to spawn new versions of itself locally on the machine. This should avoid scanners only using checksums. Sorta like a biological virus which can mutate to propagate itself. Has that been done before?

Anyway, add me to the mailing list if you're still doing this vershun@gmail.com .

0

Share this post


Link to post
Share on other sites

I would like to learn more about this project and help contribute whenever possible. Please add me to you mailing list. My email is cyclone771 [at] gmain [dot] com.

0

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0