Opinions expressed by Entrepreneur contributors are their own.
Cybercrime is on the rise in the United States and around the world. Over the past decade, online crime has cost businesses more than $12 billion, a 12-fold increase since 2015. Whereas businesses once feared physical threats like bank robberies, they are now increasingly wary of attacks that are much more invisible.
As the world becomes increasingly reliant on digital technology, artificial intelligence (AI) has the potential to strengthen cybersecurity efforts for businesses of all sizes.
Related: 10 Benefits of Using AI in Cybersecurity Practices
The role of AI in cybersecurity
Cybersecurity has changed dramatically over the past decade. Ten years ago, more than one in five cybersecurity incidents involved paper records. This year, that figure has dropped to a nearly negligible 2%. Paper record-related cyberattacks have been replaced by phishing incidents, which have nearly doubled during this time.
The rise in ransomware attacks is even more pronounced: virtually unknown in 2015, ransomware attacks now account for nearly one-third of cyberattacks in the United States. AI now plays a dual role in cybersecurity, as it is used by both perpetrators of online crimes and those seeking to protect their data. The World Economic Forum has spoken of an AI arms race.
Chief Information Security Officers (CISOs) around the world are leveraging technologies such as machine learning and deep learning to stay several steps ahead of their adversaries. Thanks to its fundamental ability to analyze vast amounts of data, AI empowers CISOs and their teams to monitor network activity and spot anomalies early on. Not only does it make it easier to identify suspicious or fraudulent behavior early on, it could be the key to effectively prevent financial and other damages.
At the same time, AI can free up analysts' time to focus on more strategic activities.
Key AI technologies in cybersecurity
Threat detection and data protection are two cornerstones of effective cybersecurity, and AI can enhance both.
Amazon GuardDuty is an AI-based threat detection capability that uses machine learning (ML) and Amazon Web Services' (AWS) unique integrated threat intelligence to protect your AWS accounts and associated data. The software detects suspicious activity, such as the removal of AI security guardrails, and investigates it faster than a human analyst could. It also allows users to automate remediation of threats.
Additionally, Amazon GuardDuty protects against ransomware uploads and scans for the presence of malware, including cryptocurrency-related threats.
IBM Watson for Cybersecurity is another powerful tool in AI-based threat detection. Watson analyzes security data from a variety of sources, including security alerts and logs. By combining information collected in multiple places, Watson can detect threats that non-AI security software may miss. The company's range of IBM Security® solutions offers more tailored tools to meet the cybersecurity needs of today's enterprises.
Related: How businesses can leverage AI and quantum technology to improve cybersecurity
Challenges and limitations
One of the challenges of using AI in cybersecurity is the speed at which the field is expanding: Research firm Gartner predicts that by 2026, four in five companies will be using generative AI-enabled applications. By comparison, just a year ago, only 5% of companies had adopted these emerging technologies.
Given the accelerating pace, one of the challenges for cybersecurity professionals is securing these projects: According to IBM, currently less than a quarter of generative AI projects are properly secured, leaving three-quarters vulnerable to malicious attacks.
Cybersecurity teams face internal and external challenges. Externally, cyber attacks are becoming more sophisticated as the volume and value of data grows. Potential attackers no longer need excessive computing power to develop harmful products.
At the same time, teams trying to protect their organization's data are dealing with internal challenges like huge data volumes and increasingly complex data infrastructures, and they must balance user access needs with security concerns.
Case Study
Cyberattacks are not limited to any particular industry. As companies become more dependent on data, so does the potential for cybercrime. Industrial solutions provider Andritz AG saw an increase in security breaches in 2020. With nearly 50% of its 27,000 employees relying on remote network access, the company needed to find a solution to protect its proprietary information and enable its teams to continue working.
With customers operating power plants, chemical plants, and more, a cybersecurity breach could quickly have devastating consequences. The team implemented an AI-based solution to visualize and quickly neutralize threats, protecting employees and customers.
China-based United Family Healthcare (UFH) turned to AI-based cybersecurity solutions after a combination of compliance regulations, external threats, and employee behaviors such as password sharing put their data at risk.
The company needed a solution that would streamline security operations centrally and enable its relatively small cybersecurity team to monitor and detect suspicious activity in various locations. Once online, the company's AI-based system provided an overview of the company's security status within minutes. The application has also had a positive effect on internal operations, reducing risks and alerts since its deployment four years ago.
Related: How AI is shaping the cybersecurity landscape – explore the benefits and limitations
Future Trends
As the use of generative AI continues to expand across industries and society at large, the need for equally strong cybersecurity will grow.
While it may still be easy to think that cybersecurity doesn't concern small businesses, few businesses can ignore cyber threats for long. Losing access to an e-commerce website or having customer data on social media platforms compromised could threaten the future of your company.
As cyber threats evolve and their scope expands, AI-based cybersecurity must evolve to detect and neutralize threats before damage occurs. In the future, we will see AI-based security elements built into widely used applications such as social media content generation, scheduling tools, and more.
When it comes to cybersecurity, AI plays a dual role as both adversary and protector. In the coming years, governments and businesses will need to ensure that AI-based protection, detection, and resolution outpace the rise in cyber threats and cybercrime.
