Securing Things Newsletter
Posts
Use of AI in Cybersecurity - Insights, Guidance, News and All

Use of AI in Cybersecurity - Insights, Guidance, News and All

[Securing Things by M. Yousuf Faisal]

M. Yousuf Faisal
July 24, 2024

In partnership with

Disclaimer: All views presented here, in this newsletter, are my own.

Author or the newsletter are not liable for any actions taken by any individual or any organization / business / entity. The information provided is for education and awareness purposes only and is not specific to any business and or situation.

M. Yousuf Faisal

Use of AI in Cybersecurity
My Recent Most Viewed Posts:
(Sponsored - Non AI cybersecurity)
Ways in which I can help?
My Asks

Hi Securing Things Community,

In this newsletter edition, I’ll be sharing at a very high-level the use of AI within Cybersecurity - some insights, interesting links, guides, posts etc. In addition, some of my most viewed social media posts (in case you’ve not seen), sponsors message and my asks.

In case you are here for the 1st time, I’d humbly request you to have a look at all the previous 25+ editions in which I have focused on different IT and OT cybersecurity, and AI insights, topics, concepts, quarterly reports and more.

Special Message:

Before we begin, do me a favour and make sure you hit the “Subscribe” button to let me know that you care and keep me motivated to publish more. Thanks!

Note: remember to validate your email address to ensure that you don’t miss any future newsletter editions.

Use of AI in Cybersecurity

As with any other industry or business, the proliferation and the use of AI capabilities within the Cybersecurity industry cannot be overlooked as the rising cybersecurity concerns with the use of AI and AI related cybersecurity threats.

Cybersecurity automation using AI has become critical as businesses struggle with security skills and budget shortage. Benefits that’s driving the industry is new forms of AI born cyber attack surface, threats & risks, limiting human error, cost efficiency, automation of processes and better decision making.

Industry is seeing the use of AI in almost all domains across IT & OT cybersecurity, ranging from but not limited to the following:

Governance
Risk identification, Assessment and Risk Management
Security Awareness Program
Secure Software Development
Vulnerability Management
Protection solutions (e.g. use of password EDR tools etc.)
Security Operations
- Intrusion Detection solutions
- Threat Detection, Hunting, and Intelligence
Incident Response, malware reverse engineering and forensics.
Supplement understaffed security teams.
and more…

Cybersecurity Tools Market using AI

Phil Alape, founder of Cyber Symposiums in his recent LinkedIn post and Richard Stiennon LinkedIn post highlighted the results of the survey from >3000 vendors (who have an AI component in place) across 10 market segments as below:

AI Conference & Bake-Off - Don’t miss the chance for registering for the 10 week online program for AI in Cybersecurity brought to us by Cyber Symposiums + Cloud Security Alliance on Sep 17 - Dec 18, 2024.

I’ve previously shared insights on the list of standards and guidance available that serves around the governing the use of AI. Checkout the following:

AI Cybersecurity Policy & Reference Guidance - which covers; AI cybersecurity policy construction (key elements), steps for a policy build execution project and several reference standards, frameworks and guidance documents.

OWASP notably provides key resources such as their AI security and OWASP AI Exchange, Periodic Table of AI Security, AI Security and Privacy Guide and LLM Top 10. Recently, the group released its LLM AI Cybersecurity & Governance Checklist.

OpenAI recently released a guidance on Optimizing LLMs for Accuracy, with super useful chart below. It’s a reminder for all that this process is not linear, and can reference this chart to shape what techniques to use for individual LLM use case.

Horizontal axis: LLM optimization - how the model needs to act → increase how consistent the behaviour is.

Vertical axis: Context optimization - what the model needs to know → increase accuracy of responses.

Source: Open AI blog

Google have SAIF - Secure AI Framework. SAIF is designed to address top-of-mind concerns for security professionals, such as AI/ML model risk management, security, and privacy — helping to ensure that when AI models are implemented, they are secure-by-default.

ISACA have released an AI Governance Toolkit - it’s a paid resource but you can find some description here. The toolkit is a list of checks for assessing governance and controls over the AI system in an enterprise. It provide assistance to auditors in verifying that AI systems adhere to the highest standards of governance and ethical responsibility.

There’s certainly a lot that I didn’t cover above. I’ll be populating the list below in future, so make sure to bookmark this.

Comment below to add anything critical that I have not yet included.

Here’s a Comparison of Top 25 AI Governance Tools - Vendor Benchmark in 2024.

31 Companies to Know in AI Cybersecurity - by Built-In written by Alyssa Schroer covers a list of companies using AI in their cybersecurity tools / solutions offerings.