In celebration of June’s London Tech Week 2019, we examined the future of technology and how it will affect our world.
In this article, we shall be discussing how AI can help combat cyber threats.
Cybersecurity tops the list of business leaders’ concerns according to research from the World Economic Forum. The survey, ‘Regional Risks for Doing Business 2018’, asked 12,000 professionals from 140 countries what their biggest concern was for their businesses. Amongst the world’s most advanced economies, Europe, Asia (China, Japan and Korea) and North America, cybersecurity polled as the number one concern. Cyber attacks are becoming more sophisticated, meaning that investment to stop them is increasing. In 2017, over $101 billion was invested globally into the sector.
Most businesses today are made up of a digital infrastructure that underpins their entire organisation. Data centres, websites, programmes, servers or accounts can all be exploited through a cyber attack. Cybercriminals are made up of a vast, highly organised groups, increasing both the severity and frequency of cyber threats. It is therefore vital that businesses utilise effective approaches to avoid being at risk.
One method that is becoming increasingly significant in protecting people and businesses from cyber attacks is artificial intelligence (AI), which is growing ever more advanced. AI’s contribution to the fight against cyber threats is that it enables analysts to respond to them with greater speed and efficiency. For example, AI can identify the patterns and behaviour of malware using algorithms and codes. This use of predictive analytics is far faster than a manual approach in monitoring and reducing the risk of cyber threats. AI can also change, develop and adapt so that it can keep pace with the scale of the threats it faces. It will remember what it has learned, resulting in an automatic response.
With this in mind, how can AI be used to protect businesses against cyber attacks?
Using Biometric Logins
There have been several highly publicised security breaches within large corporations, such as Amazon. Days before Black Friday, names and email addresses of the company’s customers were released onto the website. This was due, as Amazon stated, to a technical issue and not a security breach. It informed its customers that there would be no need for them to change their passwords. Richard Walters, Chief Technical Officer of cybersecurity firm CensorNet, disagreed.
When an instance such as this occurs, it is prudent for customers to change any passwords that are associated with a company that has suffered a technical error or security breach. This is because some passwords are available on the dark web due to previous high profile breaches. To prevent the theft of information such as passwords, biometric logins are becoming more prominent. These can be applied to fingerprints, retinal scans and palm prints, as seen with some smartphones that can be unlocked with the user’s fingerprint.
Learning with Natural Language Processing
AI-powered systems can now collect data by scanning articles, news and studies about cyber attacks. Using Natural Language Processing, they can select keywords and information from the scanned text and pick out key points. This faster method of gathering information allows cybersecurity firms to efficiently identify risks and timescales, as well as make predictions about potential threats in the future.
Securing Conditional Access
AI can help create authentication that can alter access privileges based on the user’s location or network. This makes authentication models more agile and helps to protect data from being accessed by unwanted parties. Multi-Factor Authentication collects information with the purpose of learning the behaviour of the user, application, device, network, data and location. This form of AI can then automatically change any user’s access privileges to ensure data security on remote networks.
Cybersecurity depends on three pillars. These include technology and processes, but also people. AI offers a technological solution to lessen the threat of cyber attacks, but even with automation, there still needs to be human monitoring. Even though people cannot keep pace with the malicious technology that is developing at such great speed, automated software is not immune to making errors. Security analysts can use their intuition in preventing cyber threats, but AI acts as their trusted advisor, allowing them to access more intelligence to bolster their organisation’s cybersecurity.
By taking a modern approach of combining human insight with AI that can adapt and evolve as cyberthreats develop, businesses stand a far better chance of avoiding attacks upon their digital infrastructure.