THE BLOG

How AI Might Transform The Landscape Of Cybersecurity

22/06/2017 15:36 BST | Updated 22/06/2017 15:36 BST

2017-06-20-1497962706-7085536-shutterstock_579200824.jpg

Image source: Shutterstock

The growth of newer technologies like IoT, cloud computing, and smartphones have greatly contributed to the evolving sector of cyber security but have also sparked concerns when it comes to security risks associated with the use of these technologies. Cyber criminals now have a plethora of access points and connected devices to target and with innovative and smarter ways to do so at their helm, mitigating these threats deeply overshadows the concept of "bringing innovation" to these technologies.

Businesses particularly, have leveraged these technologies owing to the fact that they have greatly added to the convenience of their operations and automation. However, with the gaudy headlines of mega breaches and enormous DDoS attacks we come across every day, the desire to implement these technologies is put on a back burner as those with limited financial muscle, can't afford to be another delicacy at the dinner table of cyber-criminals. With this scenario in mind, the question still remains - can businesses opt for newer technologies like AI and remain protected at the same time?

Artificial Intelligence (AI) - another feast for cyber criminals?

That future is not far from reality where businesses would be exploiting the full potential of artificial intelligence (AI) to complement their data collection, data scrutiny & analysis, keeping a tap on new trends, and, protection of valuable customer data. But, does that mean with smarter and capable ways, businesses would have to become unwary of bucking up their cyber security infrastructure? The simple answer is a big "nod".

As processes become more capable and with thorough integration of AI in businesses, we will begin to witness more sophisticated threats which would be designed to penetrate stronger networks, carry out personal data theft and, spreading intelligent viruses and ransomware. Moreover, with an imminent all-out coming of AI, numerous security problems might plague the already fragile cyber security infrastructure. One of them is "intentional malice" where malicious task intelligence or goal might be introduced in the initial stage of AI integration or with an AI behaviour modifying software in the later stage, it can be directed to act in a suspicious way or might be programmed to inflict damage on its own.

Alarmingly, some of the greatest minds including Bill Gates, Stephen Hawking and Elon Musk have also voiced their concerns on the repercussions AI would bring with it and how it has the ability of directing itself and getting out of human control. Imagine a world where everybody would be in the close proximity of AI control mechanism and one move can "alter or destroy human race".

Will the threats get smarter?

In the coming years, we might see attacks getting smarter - having the ability to operate autonomously without the need of human interference. AI-controlled smart malware is one such example, a malware which can have adaptive learning mechanism to improve the effectiveness and penetrability of the attacks. This includes understanding the environment, making calculated decisions based on that environment, exploration of potential targets and the worst case scenario - intelligently evading all sorts of detection.

These smart manoeuvring autonomous threats might be able to make their way to different platforms, be it mobile based systems, personal computers or IoT, but that would mostly be depending on the level of security these devices have or they would be exploiting the weakest link in the system. IoT in this case might paint a gruelling picture as the security infrastructure present on these devices is still by and large, susceptible to cyber-attacks and with AI kicking in with its open claws not so long from now, it is sure to take on the most catastrophic blow.

AI as a close cyber security accomplice

Apart from the above mentioned horrendous scenario, AI can also spell wonders for the organisations looking to beef up their cyber defences with the help of hybrid approach involving human-AI interaction on a broader scale. With the help of predictive analytics, security experts would then be spending their precious energies on making strategies to cope up with cyber threats based on the insights provided by AI rather than

Additionally, with the right set of tasks and its proper distribution, we can always remain one step ahead of AI. An ideal human-AI partnership will usher a new era of threat detection and will help us in designing impenetrable security systems. A recent report by UK's government office for science sees contribution of AI in fourth industrial revolution and changes it would bring to traditional skillset.

A dark side - The duo of AI and cyber criminals

Imagine a scenario, where hackers will start experimenting with AI to target their next victim through stolen data. Spear phishing is the prime example in this case. Instead of skimming in the pile of data to target a single victim, hackers would be using AI to smartly analyze the digital footprints of its victim, finding the victims with the most loose ends and carrying out closely calculated attacks, that too with a minimal chance to fail.

According to Marc Goodman, author of Future Crimes and a law enforcement adviser by profession, "AI has the potential of being used to create cyber weapons and people are still unaware about the fact that cybercrime is becoming highly automated and growing rapidly".

Final words

While cybercriminals quickly learn to use emerging technologies for their own benefit, it's the most ripe time to delve into the thought process on how artificial intelligence will fit into cyber security infrastructure of businesses and with AI still in the early stages, it's time to work on strategies to utilise it fully once it marks its second coming in the business world.