Machine learning and AI

Is AI really the Future of Cyber Security

November 6, 2017 by Irfan Shakeel

As we progressively get more dependent on technological innovation in our daily lives, we open ourselves up to cyber-attacks. Each device that we use today is vulnerable to cyber-attacks. Hackers are rapidly evolving to come up with new strategies and attack to breach our systems, steal our information or sabotage it for ransom drives.

Cybersecurity risks grow alongside technological progress. Till 2016, Hackers have taken a proactive and innovative approach to developing strategies to infiltrate our systems. Now, it is our turn to show them that we are well prepared to defeat their strategies and safeguard our information system more effectively than ever before.

However, employers are looking further outside of IT for tech talent. Organizations are not just looking for the standard computer engineer anymore. While they still need engineers, developers, data scientists and technological tools to write, pull and track data; the need to have professionals who can make sense of all of that information and communicate it back to the executive team is becoming increasingly more important. To reduce the risk of a cyber-threat the IT professionals in your origination need to work as a team.

We have witnessed various strategies that are adopted at different levels to prevent hackers or malicious application to sneak into our networks. To assure the security of information, we have two-way authentication, rule base block list, network layer security, end-point security, security tools to track user activities and many another type of security rules and policies that are helping in preventing malicious files or actors to harm our information systems.

To address future security concerns like cloud security, IoT security and enhanced device usage in future, the IT world is looking forward to adopting Artificial Intelligence-based security tools and applications that can provide real-time protection and take decisions on its own to block suspicious user or file to propagate through the network. Moreover, as artificial intelligence (AI) systems become more capable, we will begin to see more automated and increasingly sophisticated social engineering attacks. The rise of AI-enabled cyber-attacks is expected to cause an explosion of network penetrations, personal data thefts, and an epidemic-level spread of intelligent computer viruses. Ironically, our best hope to defend against AI-enabled hacking is by using AI.

In an interview, Ondrej Vlcek, CTO & GM of Consumer at Avast said, “In 10-15 years, we will be deep in a ‘war of the machines’ era with advances in artificial intelligence bringing fast and sophisticated execution of security defense and cybercrime. This will be a battle of AI vs. AI. The availability of low-cost computing and storage, off-the-shelf machine learning algorithms, AI code and open AI platforms will drive increased AI use by the good guys to defend and protect – but also increase deployment of AI by the bad guys. There will be sophisticated attacks launched on a grand scale, quickly and intelligently with little human intervention that compromise our digital devices and web infrastructure.”

Indeed, AI can be utilized to shield and to attack digital infrastructure, and in addition to building the attack surface that programmers can focus on, that is, the quantity of courses for programmers to get into a framework. Ondrej further added, “Cybercriminals will create fully autonomous, AI-based attacks that will operate completely independently adapt, make decisions on their own and more. Security companies will counter this by developing and deploying AI-based defensive systems.”

However, a lot of AI development is being spent in the cybersecurity space; as well it should with the dawn of ransomware, refined and enhanced malware and other. All the top technology companies are spending millions each year on AI and cybersecurity — from Microsoft to Google, Cisco, Symantec and including the big name anti-virus firms. However, in the last few years, there has been an increase in startups around security tools as well that tout machine learning and AI (Darktrace, Cylance, AlienVault, etc.).

AI malware:

Imagine you have a meeting with a customer, and in a matter of seconds, before you leave, they send you over a confirmation and a guide with directions to where you intend to meet. Everything looks ordinary — however, the whole message was really composed by an AI-powered malware imitating the customer’s email characteristics, with a virus appended to the guide.

By utilizing recurring neural networks, it is as of now conceivable to show AI programming to emulate composing styles. Moreover, AI-infused ransomware could turbo-charge the dangers these attacks make — self-organizing out to inflict maximum harm, and following new, significantly more lucrative targets.

While there is no “silver bullet” when it comes to protecting your company’s network, it is important to have a robust, multi-layered security strategy that is backed up by the AI algorithm to take decisions on its own. Unfortunately, those who are becoming more advanced when it comes to AI regarding security are the ones on the offensive (cybercriminals). The way to fight against these criminals is to escalate AI defenses.

Moreover, on top of your traditional firewall and IPS (Intrusion Prevention System), organizations should consider adding an industry-proven endpoint monitoring system, preferably one that uses machine learning to identify and prevent bad code from executing. Then, adding a tool that gives a holistic view of the entire network in real time that identifies advanced threats, including those stealthy, unconventional, silent attackers would make it more challenging for cybercriminals to sneak into organization’s network.

Posted: November 6, 2017
Irfan Shakeel
View Profile

Irfan Shakeel is the founder & CEO of An engineer, penetration tester and a security researcher. He specializes in Network, VoIP Penetration testing and digital forensics. He is the author of the book title “Hacking from Scratch”. He loves to provide training and consultancy services, and working as an independent security researcher.