Cyber security & AI for SaaS

[Nicolayyyy]

Hi there,

Last night I thought about the combination of AI and cyber security to make a SaaS. With the rising importance of cybersecurity, integrating AI could help to prevent threats and scan companies networks for recommendation to make.
While there are existing tools in the market, I believe there's always room for innovation...

I'm currently pursuing a degree in cybersecurity and will graduate in a year. But I have no idea about how could I make a solution like that, better than those which already exist, and I don't know how AI and machine learning works and their integration with cybersecurity. I guess it'll take a while to learn and build a solution like this, and it could be a good long-term investment but the potential could be boundless.

I'd love to hear your thoughts on this topic!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[ZackerySprague]

The best way is to get a job in Cyber and find needs you have tools already such as Rapid7 Nessus, Qualls, SentielOne.

You'd have to find ways to be very different or find different value skews that the current software that's on the market. AI is becoming super popular now days and will mostly likely have some sort integration or feature into the current market products.

I'm not sure I'd even want to tackle the market of cybsecurity and AI.

A friend from this forum told threw some ideas on how AI could be used in different markets instead.

To me it seems that the space will become very competitive, very fast.

If anything speak to your audience and what problems they have. Another way that I found a problem was from the job I has at the beginning of this month to which I now regret quitting, but I found a need that could be solved and will have to create it and test it in the marketplace.

This field though is starting to frustrate me, so much hurt, in a short amount of time. I don't mind lending a helping hand to give you a head start when you get out. Shoot me a message and I don't mind getting on a quick call. Getting into Cyber is becoming difficult.

IT has always been the "Holy Grail" of jobs that was advertised as secure, but its not. So don't be mislead by the titles, but don't be scared either. It's an ever evolving field that requires techs to keep their skills up to date.

Don't be siloed into a role that doesn't allow you to grow, test, and learn new skills. Always been experimenting and making mistakes, acquiring certs to keep yourself up to date.

Best of all is just to get into the Arena of a Cyber job, find pain points, call the potential companies to see if they would buy it. A small soft validation if you will, then build the MVP and acquire your first customer.

 

[ZackerySprague]

A problem I discovered was ITAD companies not tracking inventory such as laptops, desktops, monitors, etc. Their just shipping product out the door without ever adding to an inventory management system. I think the market could be disrupted though.

I even have a few great ideas that I learned from somebody on a network event for the keyboards, mouse, and parts. It's kinda cool. Kinda excited if I can bring it to life as well.

 

[ZackerySprague]

Another thing I learned from Mike Rowe from a friend is that you shouldn't identify yourself with your current profession, instead try to gain experience this could come from a job, Freelancing, or starting a company. A lot to unpack for another time.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Nicolayyyy]

Thanks for your replies.

I'm struggling to know what I want to do in cyber security. I have some ideas in other domains, with machine learnings, some apps, but I find this hard to find a solution which respects the Fast Lane process in relation to cyber security.

I'm currently a student from Rob's Fox Legend Program and I've done my first project, so I'm really happy and I'm trying to continue find some new clients to make other deals. I also thought to "upgrade" this web design stuff to a Digital Marketing company, it could be good, but in the short-term.

I have an internship to make next year before being graduated, I hope that it'll help me find some issues that I could solve..

IT has always been the "Holy Grail" of jobs that was advertised as secure, but its not. So don't be mislead by the titles, but don't be scared either. It's an ever evolving field that requires techs to keep their skills up to date.

Everybody tells me that IT jobs and cyber security jobs are "the future" and "there will always be a job for you in this field".. So I'm a bit lost about what decisions I need to make for the future, and I know that I won't be happy in that kind of environment.

This field though is starting to frustrate me, so much hurt, in a short amount of time. I don't mind lending a helping hand to give you a head start when you get out. Shoot me a message and I don't mind getting on a quick call.

Thanks, of course it'll be a pleasure to get some help. What do you do in life ?

 

[ZackerySprague]

In Cybersecurity you have three roughly generic umbrellas:

1. Penetration Testing
2. Incident Response
3. Digital Forensics
4. Compliance

If your desire is to go Fastlane then the idea, would to see the job as a stepping stone instead of a career.

It is true that you will always find a job in IT but it's sometimes never a good idea to stay in one place where it's very regimented in doing one skill over and over again.

What you have to figure out is what is your end goal? Is it to be in IT or to be an entrepreneur?

Fox's courses teaches you how to step into the world of becoming a freelancer by learning web design skills, sales skills, Prospecting, and business acumen to sell projects in a high priced range and solving problems. So higher hourly rate and less time to do the work per project. It's to help gain capital in order to fund Fastlane ventures.

I can teach you just how to get a head start into getting into Cyber to find your pain points such as what certificates to get, what degrees to get (cheaply), how to create resume, etc. Just some generic advice.

One good idea (to which I am need to learn) is to focus on one thing at the moment.

A good idea to start is to read Nathan Barry's ladders to wealth article to get you a good idea on which Ladder you are on or should focus on atm.

 

[Nicolayyyy]

Is it to be in IT or to be an entrepreneur?

My God, just the idea of being in IT terrifies me. Since I've discover through MJ DeMarco's books that it's possible, this life is not only reserved to special people with special skills, I can't see myself in the comfort life, with a job until my 60s. So yes, my goal is to be an entrepreneur and be able to be useful in this world by resolving others problems.

Fox's courses teaches you how to step into the world of becoming a freelancer by learning web design skills, sales skills, Prospecting, and business acumen to sell projects in a high priced range and solving problems. So higher hourly rate and less time to do the work per project. It's to help gain capital in order to fund Fastlane ventures.

Exactly, and I still have things to learn and take action of this learning.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[ZackerySprague]

Information Technology isn't a bad field, I love computers and tech forever and always. My dislike is the people that's in and how companies are run. Corporate America and I are not best friends, I have my reasons but I make due.

The field is ever evolving and it's excellent to be in, even for Fastlane ventures. If I am to be honest, I love anything and all tech related stuff. I've just been out of touch.

As in our Fastlane meetup today in DFW, the advice that was given which everyone should heed as well is that. You will not be in a position forever and should always been learning something new on the side or in the job. You should always been looking for the next thing you will do or venture you want to pursue. So while you are at this job always been learning.

I made this mistake by not doing it, took every paycheck for the last year up until now to squeeze and payback every debt that I could. I should have used the certificate programs (if it wasn't me paying for it and them reimbursing it) to keep up with mu skills.

If Cybersecurity and AI is what you want to do then look into markets like agriculture, cars, etc. Old stale markets that still see huge potential in growth year after year.

Micha Rohressien saw that the car market was worth $1 trillion in market cap and created the company prodigy, helped buyers buy cars online. He observed another company who got VC in a small sub-niche for helping buyers online buy cars online. Just stole the idea and made it better. I still believe he read Unscripted the 2017 version to find this idea. Because his company dates back in 2018 when it started on Archive.org. oh well.

 

[Tau Ceti]

I will tell you the same thing I said to someone who came to this forum months ago and wanted to start an analytics company(think paid Google Analytics) and add AI to it.

If your only moat is adding AI to your product, then you don't have a product. AI in this context would being GPT-4 api calls.

Now if you were to build your proprietary models with proprietary data and sell access to them to customers, then, you would have a moat of some kind.

But the question remains, why is AI needed in the first place and how is it going to make cybersecurity better? If you don't have an answer for that question, then you don't have business.

 

[EmperorPear]

Hi there,

Last night I thought about the combination of AI and cyber security to make a SaaS. With the rising importance of cybersecurity, integrating AI could help to prevent threats and scan companies networks for recommendation to make.
While there are existing tools in the market, I believe there's always room for innovation...

I'm currently pursuing a degree in cybersecurity and will graduate in a year. But I have no idea about how could I make a solution like that, better than those which already exist, and I don't know how AI and machine learning works and their integration with cybersecurity. I guess it'll take a while to learn and build a solution like this, and it could be a good long-term investment but the potential could be boundless.

I'd love to hear your thoughts on this topic!

The first step seems to be to see what the current solutions are and try to find a flaw in them. Since you're a cybersec expert, if anyone could find a flaw, it should be you. Once you find something lacking, then it's time to make an improved solution.

From a purely technical standpoint, I would go with a transaction analysis system by using Kafka for real-time data processing, and use AI to analyzed the data. Usually, attack detection is done by using rule-based restrictions, threshold-based restrictions, or both. A machine learning model or neural network would come into play with threshold-based restrictions.

Now, since you said you don't know how to develop AI, that sounds like a good place to start, because this is not something ChatGPT will do for you. Without going too much into technical detail, a neural network uses numerical parameters to determine an output from your input, ie. like a function. It also uses equations and algorithms like backpropagation or convex optimization (this one is more commonly known as deep learning) to adjust for errors in output by changing the parameters themselves, thereby learning from its mistakes.

Now that the quick rundown has been given, I can explain why you should also secure your neural network as well! Now, say you have successfully trained your AI and you have your parameters. If these were to be discovered you would face issues ranging from attackers circumventing your AI, or competitors copying your parameters for their own product! As for how you should encrypt an AI's parameters, I would suggest homomorphic encryption, which allows you to perform computations on encrypted data. This is already used for certain neural networks in the medical field, in order to build accurate diagnostic networks, while preserving patient confidentiality.

It is an interesting project to make, and I hope I've given you some ways to approach your problem and some topics to think about as you begin to solve it, rather than just bog you down with overly technical stuff Good luck!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.