Cybersecurity and Generative AI? Here’s a Look at Why Both Technologies are a Perfect Fit

Cybersecurity is an important issue for businesses across industries. As most companies set up an online presence to tap into a bigger market, the risks also increase.  Cyber threats are common, and they can be costly and disruptive. In 2023 alone, experts have tracked over 2,300 cyberattacks, representing over 343 million victims. The data indicates a 72% increase in data breaches since 2021, the previous record, with phishing, malware, and ransomware as the most widespread threats.

Cybercrime surges as the world becomes more interconnected and dependent on new technologies. Even with this new technology, led by generative AI, cybercrime remains a larger-than-life threat.

The industry was rattled when OpenAI reported that its flagship project, ChatGPT, was hit by a data breach. Some have ventured into saying that AI is a looming cyber-security threat. For example, Microsoft has reported that some hackers from Iran, North Korea, and Russia used Open AI to carry out cyber attacks, which they managed to stop. These types of attacks, according to experts, are no longer surprising since generative AI can automate phishing attacks. It’s now easier for bad actors to design phishing attacks by scraping information from the web, matching these details, and associating them with a specific person.

But this doesn’t mean we’re at the mercy of new technologies. We have witnessed how cyber threats can disrupt our workflows. However, new technologies can also help, especially in setting up a reliable cybersecurity program that fits the times.

Generative AI is now powering many cybersecurity programs

Did you know that generative AI is a perfect fit for cybersecurity programs? Just check top tech and consultancy firms' latest press and product releases, and you’ll know how fast generative AI is being integrated into online security programs. IBM, one of the leading online security companies, has already announced the integration of generative AI into its managed Threat Detection and Response Services. 

IBM Consulting analysts now use this service to streamline security services for its customers. According to a corporate press release, these new services are built on the company’s watsonx data and AI platform and, once deployed, can help automate and improve the identification, investigation, and response to security threats.

Even small businesses and organizations can use generative AI to boost their cybersecurity strategies. Here are a few real-life applications of AI for security:

• Integration of AI into the company’s threat detection systems. Generative AI can help manage and protect the company’s email delivery chain. Traditionally, businesses choose to detect email threats before or after the emails are delivered. However, detecting threats pre-delivery, post-delivery, and even during click-time is impossible for some companies.
• It helps identify patterns suggestive of cyber threats. Generative AI can also help security operations centers (SOCs) identify patterns of potential cyber threats like ransomware, malware, or unusual spikes in traffic. The technology offers more sophisticated data analysis and detection of network abnormalities. Generative AI tools can learn from historical security data, establish a baseline ‘normal behavior,’ and then flag deviations indicative of security incidents.
• AI helps reduce human bias. According to a National Technical Information Service data scientist, AI can support decision-making and reduce human bias. For example, the Department of Justice can use technology to analyze case law by considering demographic data and historical sentencing information to develop objective sentencing recommendations. This approach can also extend to designing a data-rich cyber security program with reduced human bias.
• AI can analyze large-scale sensitive data and identify associated risks. Companies can also leverage AI technology to understand and locate sensitive data, identify who has access to it, and identify the inherent risks. AI can connect all information to assess data access, vulnerability, and sensitivity, giving owners a more intelligent approach to planning and crafting a cybersecurity plan.

It’s time to appreciate generative AI’s increasing role in cybersecurity

For many cybersecurity experts, AI is no longer just a competitive advantage; it’s now a requirement for any business.

Businesses now face a complex and evolving threat landscape defined by sophisticated online activities. Phishing is becoming more aggressive, with increasing cases of identity theft. Owners and analysts require advanced tools to keep pace with these threats. As mentioned in this article, AI enables companies to manage and mitigate security threats more effectively, offering some form of automation, objectivity, and insights that aren’t found in traditional methods. As large language models or LLMs are pre-trained on vast amounts of data, business can quickly scale their cybersecurity strategies depending on the size of operations and the amount of data.

Key AI security policy considerations

Given the increasing complexity and aggressiveness of online threats, promoting privacy and protecting data become the primary objectives of any AI security policy. With these things in mind, any cybersecurity plan should consider the following:

• Don’t share private information with public AI platforms or tools that the company doesn’t control. Employees and workers must be reminded not to share corporate information with free generative AI tools like ChatGPT.
• Adopt a policy separating data. Businesses must adopt a policy that prevents the combination of different kinds of data and their sharing with the public. For example, companies can adopt a data classification system.
• Learn to verify and fact-check information from AI platforms. ChatGPT and other chatbots are becoming more popular and used for research and marketing. However, these tools feed on available information online, and since there is plenty of fake and biased information, there’s a potential to generate false and misleading content. As such, businesses must learn how to vet and fact-check information generated by these models.
• Zero-trust policy. Businesses must adopt a zero-trust posture as a way to manage risks.

Although AI advancements benefit many industries and boost security, experts say it isn’t a completely reliable tool. AI isn’t a complete replacement for human analysis, intervention, and recommendations. Instead, businesses should acknowledge the technology as a tool to augment existing security capabilities, allowing them to address the fast-changing security threat landscape confidently.