Micki Boland is a global cyber security warrior and evangelist with Check Point Technologies’ Office of the CTO. Micki has over 20 years in ICT, cyber security, emerging technology and innovation. Micki’s focus is helping customers, system integrators, and service providers reduce risk through the adoption of emerging cyber security technologies. Micki is an ISC2 CISSP and holds a Master of Science in Technology Commercialization from the University of Texas at Austin, and an MBA with a global security concentration from East Carolina University.
In this two part interview series, Micki Boland will discuss Artificial Intelligence (AI) and Machine Learning (ML) as it pertains to cyber security and modern business challenges. Did you miss part 1? Click here.
Which industries are best suited to AI/ML adoption (if any)?
Wherever there is data and decisions are to be made, there is AI. We use AI in just about every industry. One of the most exciting areas of AI use is in the healthcare industry, which is exquisitely suited to AI/ML and deep learning using neural networks; healthcare can apply AI/ML and deep learning everywhere, from bioinformatics to patient care, to epidemiology, to pharmaceutical development and real world life saving medical advances. There are so many amazing applications for AI.
Artificial neural networks (ANN) are working under the hood as we speak to combat the coronavirus pandemic. ANN is the power behind rapid identification of the specific antibody immune response to coronavirus that has enabled the rapid development of coronavirus vaccines.
What are the biggest payoffs for the industries?
AI technology can provide big payoffs in terms of the agility, speed, innovation and the competitiveness needed to deal with complexity in every industry sector. The demand for AI technology is huge. Consider the March 2020 IEEE Spectrum snapshot of the top 40 United States based AI startups, and you will find valuations ranging from $100m to $1b USD. Of these AI startups with valuations greater than $1b are in compute/software, healthcare, finance, automation tools, and biological and agriculture industry sectors.
Actual payoffs are sometimes difficult to quantify. Does AI deliver real payoffs in terms of cost savings, increased revenue, and new markets, or is it really providing incremental innovation and shifting of resources and dollars from one area of the enterprise to another? Definitely, there is a huge payoff for cloud providers! For the enterprise to gain data science expertise and empower the organization’s AI goals, organizations need to invest in data science and data science architectures. For many, this investment hurdle is too high, and this is why AI/ML/DL/ANN cloud platforms, services and technologies are being rapidly adopted as solutions to big data, data lakes, data analytics, and indeed off-the-shelf AI/ML solutions.
Can AI/ML adoption reduce overall cyber security costs?
Exposure time is the ultimate enemy of cyber security. The most important thing to do is stop the security event or breach. In the event of a cyber security breach or incident, that exposure window is wide open while IR/IH teams race to detect, investigate, determine bad guy tactics, tools, procedures (TTP), trace to corporate assets (what is exposed, what has been taken), AND then take actions to actually mitigate the threat.
In cyber security, AI/ML can dramatically reduce cyber security risk and resource costs associated with reactionary response to threats and malware. Using AI/ML based cyber security to proactively and in real time identify and mitigate threats, identify and mitigate malware, including new malware variants, is huge! It avoids huge costs in regulatory fines, legal costs, and in public relations when it comes to dealing with cyber security events and data breaches. Some costs are less tangible, including negative reputation impact and customer churn after publicly reported cyber security events. The enterprise gets the best of cyber security AI/ML and this enables the enterprise to reduce cyber risk and to focus on its core business.
What should organizations look for in obtaining cyber security architectures that contain AI/ML?
Utilizing deep learning for anomaly detection provides the capability of detecting normal and abnormal traffic. DL needs to be able to understand, define, and integrate with the entire infrastructure including network, compute, storage, endpoints, applications, and databases. Cyber security enforcement points, agents, nano agents, and gateways enable continuous learning about the infrastructure and permits DL anomaly detection. An integrated approach is leveraging an architecture framework like Check Point Software Technologies’ Infinity architecture. This framework is adaptive and can help organizations achieve the goals of AI-based cyber security. This means having a comprehensive cyber security solution provider that is delivering cyber security AI/ML for the organization. In the case of Check Point Software Technologies, we have a data lake of real time, contextual threat intelligence called ThreatCloud and our enforcement points (networks, endpoints, cloud, mobile, physical, IoT, ICS/SCADA) can consume this threat intelligence and automatically invoke protection actions. We have a lot of data for each decision point in our products, and have the best experts in the world to correctly label and enrich this data. With both resources, the AI that we introduce into the decision points in our products is not magic – it is a reflection of the best experts’ thought processes on vast, relevant data, packed in a box for our customers.
Any other key takeaways that you’d like to share with the Cyber Talk audience:
If you love learning about AI, and want to geek out some more, maybe engage with your kids to get involved in some citizen AI, check out how machine learning is being used in space and how NASA and citizen scientists are using ML to identify and label objects in space. The NASA citizen projects are Galaxy Zoo, Planet Hunters, and Disk Detective. More about NASA’s Planet Hunters project here.
