[box type=”info” align=”” class=”” width=””]Dmitry Samartsev Chief Executive Officer, BI.ZONE
- Cybersecurity issues are widely seen as the greatest risk to company growth in the next three years.
- There is a global shortage of suitably trained and qualified people to work in cybersecurity.
- Artificial Intelligence (AI) could improve cybersecurity by identifying suspicious behaviour, predicting cyberattacks and keeping data secure.
According to the World Economic Forum Global Cybersecurity Outlook 2022 report, 48% of executives believe that artificial intelligence (AI) will influence cyber transformation in the next two years. AI is a powerful tool for both cybercriminals and cybersecurity experts. Hackers are using AI to make their attacks more sophisticated and harder to detect, while cybersecurity specialists are finding ways of integrating AI into corporate cybersecurity systems to minimize financial and reputational losses.
A survey of the CEOs of the world`s 500 most influential companies across 11 industries found that cybersecurity issues will be the greatest risk to company growth in the next three years. This is unsurprising, given the broader threat landscape and the global lack of cybersecurity staff, as well as the increasing frequency and severity of cyber attacks. In the face of such challenges, AI could be a game-changer.
Here’s how AI could improve cybersecurity:
Mitigating cyber threats
AI technology could help businesses to detect cyber attacks, by identifying vulnerabilities and suspicious behaviour. Currently 91% of cyberattacks do not generate an alert, and companies take on average 207 days to identify and 73 days to contain a breach. AI can analyse events rapidly, helping cybersecurity teams to keep pace with risks and associated financial and reputational losses.
AI can help to predict cyber threats. For example, financial institutions have anti-fraud systems that use algorithms to scan through large pools of data, identify patterns of typical behaviour, and make predictions of potential fraud. Such insights can feed directly into the development of systems to prevent cyber attacks. These systems are also becoming a must-have for marketplaces, airlines and all companies with customer loyalty programs.
Finally, AI can help to support under-resourced security teams. One of the biggest challenges in cybersecurity is a shortage of suitably trained and qualified workers, estimated at 2.72 million globally. According to a recent study, 57% of organizations were affected by this in 2021. As cyber risks multiply, these understaffed teams often face high levels of stress, which in turn affects their ability to react and respond to threats.
There are multiple reasons for the shortage, including high levels of stress experienced by understaffed cybersecurity teams. AI can help by using algorithms to monitor incidents across the system and flag cyber threats that require immediate action. Businesses can also use the technology to speed up the creation of virtual patches and devise appropriate responses to cyber attacks.
In the period 2020-2021, the cost of data breaches increased by 10% from $3.86 million to $4.24 million. Aside from the high financial penalties, such breaches put businesses and their customers at risk, and raise serious legal issues and implications. Organizations that have security AI and automation in place experience lower data breach costs in comparison to those that do not.
Organizations can use AI technology to strengthen their data governance process. For example, AI can sift through a variety of sources to identify sensitive data which businesses can then anonymize, remove, encrypt or protect in other ways from any external or internal threats. Such uses of data allow businesses to handle confidential information securely and comply with privacy standards and regulations such as the GDPR. Similarly, AI can create synthetic data, to create scenarios and forecast future cyber exposures.
AI-powered biometric authentication
AI is also taking biometrics to a new level. For example, facial recognition metrics always vary – there can be a different angle of head tilt, different distance between features, lighting, or facial expression. The same applies to voice recognition and other biometrics. It is impossible to set fixed rules of recognition, so AI analyzes these deviations of metrics within the specified ranges, makes decisions accordingly and learns continously. AI can also help to identify users and differentiate them from bots and automatic attacks, by analysing unique behavioral biometrics such as keystroke, scroll and swipe patterns.
As we look ahead, AI could open even more opportunities to businesses in terms of security. To resist cyberthreats successfully, businesses need to be one step ahead of the criminals.
[box type=”note” align=”” class=”” width=””]License and Republishing