"Balancing security, performance, and usability is key… We will safely support enterprises' AI adoption through AI integrated governance"

CloudNetworks (CEO Jeong Heung-gyun) announced that it continued its growth momentum in 2025, setting a new all-time high in revenue for the second consecutive year. The company explained that this was a meaningful year in which it secured both profitability and growth simultaneously, driven especially by significant growth in operating profit. Based on these achievements, it added that it expanded its organization and increased investment in strengthening its technical expertise.

In the cloud security domain, the company said that the "Cloud Zero Trust Integrated Security Model" implementation project, centered on Splunk, HashiCorp, and Qualys, progressed actively. It explained that the integrated security architecture that had been introduced through Daily Secu expanded into actual implementation cases. In the financial sector, the company said it strengthened its real-time monitoring system through a cloud traffic analysis system based on its own solutions. In the network infrastructure area as well, the company said that the scale of network system implementation projects expanded significantly, centered on the e-commerce, manufacturing, and healthcare industries. Through this, the company continued its growth momentum across the entire cloud security and network infrastructure domains, and emphasized that it is positioning itself as a company equipped with integrated technology capabilities spanning security, cloud, network, and AI.

CEO Jeong Heung-gyun cited "the full-scale arrival of an era in which AI leads attacks" as the biggest threat for 2026. He assessed that with real-time deepfake impersonation of executives and malware that evades detection by self-modifying its code, the speed and precision of attacks have increased, shortening the golden time for response, and that the security paradigm has shifted into an all-out "AI versus AI" war. Accordingly, he stated that it is necessary to build an "AI Integrated Governance" framework that encompasses AI guardrails, AI DLP, and observability, as well as safety, transparency, operational monitoring, and cost optimization.

The following is an interview with CEO Jeong Heung-gyun of CloudNetworks.

Q1. Please introduce CloudNetworks' key achievements in 2025.

CloudNetworks continued its sustained growth momentum in 2025, setting a new all-time high in revenue for the second consecutive year. In particular, it was a meaningful year in which we secured both profitability and growth simultaneously, driven by significant growth in operating profit. Based on these achievements, we expanded our organization and invested in strengthening our technical expertise.

Looking at it by business area, first, we achieved major results in the cloud security domain. The Cloud Zero Trust Integrated Security Model implementation project, centered on Splunk, HashiCorp, and Qualys, progressed actively. This was an actual implementation case of the integrated security architecture that we had introduced multiple times through Daily Secu. In addition, in the financial sector, we strengthened our real-time monitoring system through a cloud traffic analysis system utilizing our own solutions. We also achieved notable growth in the network infrastructure area. As the scale of network system implementation projects expanded significantly, centered on the e-commerce, manufacturing, and healthcare industries, we continued our growth momentum across the entire cloud security and network infrastructure domains.

Based on these achievements, CloudNetworks is positioning itself as a company equipped with integrated technology capabilities spanning security, cloud, network, and AI.

Q2. Which product lines received the best market response in 2025?

In 2025, we achieved notable results largely in three product lines.

First was the large-scale integrated log and security system implementation using Splunk. This included the largest-scale security monitoring system implementation case in Korea, building a system that centrally manages and analyzes logs and security events across the entire hybrid cloud environment, and realizing company-wide security enhancement including automated threat response.

Second was HashiCorp Terraform and Vault. As enterprises' cloud usage surged and AI infrastructure buildout began in earnest, there was strong demand to standardize and automate infrastructure deployment pipelines. Through IaC-based infrastructure automation using Terraform, we implemented a system that can respond agilely to the rapidly changing AI-cloud environment, and through secrets management using Vault, we fundamentally blocked the exposure of sensitive information through dynamic secrets management in Kubernetes environments.

Third, in the network infrastructure area, "HPE Aruba Networking" stood out. In a large-scale integrated infrastructure network implementation project, we configured an SDA network using Aruba's wired/wireless controllers and ClearPass. By automatically applying fine-grained policies based on infrastructure characteristics, users, and devices across the entire wired and wireless segments, we were able to greatly strengthen security while dramatically reducing operational complexity and incident response time.

Q3. What security threats are you most concerned about in 2026, and how do you plan to respond?

The biggest threat in 2026 is the full-scale arrival of an era in which AI leads attacks. AI is no longer merely a tool that assists attacks; it is becoming the main driver of agentic AI-based attacks that set their own goals, find vulnerabilities, and infiltrate on their own. Attacks that impersonate senior executives in real time using deepfakes are increasing, and malware evades detection by modifying its own code. As the speed and precision of attacks increase dramatically, the golden time for security personnel to respond is shrinking, and the security paradigm has now completely shifted into an all-out "AI versus AI" war.

To respond to this, we need to systematically build AI integrated governance. With the "AI Basic Act" taking effect on January 22, 2026, making the management of high-impact AI mandatory, ensuring the safety and transparency of AI is now a necessity, not an option. Beyond strengthening security through AI guardrails and AI DLP, an integrated approach is needed that also encompasses AI performance observation, operational monitoring, and resource and cost optimization.

Q4. What areas will you focus on in your 2026 product and technology roadmap?

To respond to this threat environment, CloudNetworks plans to focus on "AI Integrated Governance" in 2026. AI Integrated Governance is a comprehensive framework that encompasses not only AI security but also AI observability, performance evaluation, and resource optimization. This is because, as enterprise adoption of generative AI accelerates, existing security frameworks have become difficult to use to control the complex data flows and unpredictable behavior of AI environments.

In the AI security domain, we plan to focus on AI guardrails, AI DLP, and observability.

First, "AI guardrails" block AI model hallucinations, bias, and prompt injection attacks. Prompt injection is an attack technique that uses manipulated queries to bypass AI's security policies, and cases of it have been surging recently. Guardrails block these risks in real time at the AI's input and output stages, ensuring that AI operates only within the bounds of a company's ethical standards and legal regulations.

"AI DLP" is also a key task. Existing DLP was based on specific patterns and keywords, but its limitations are clear in AI environments that understand context. AI DLP analyzes even the intent and context of a user's query to block sensitive information from flowing into AI training data or being sent to external LLMs. As incidents of employees leaking corporate confidential information through public AI services such as ChatGPT or Claude continue to occur, the need for this area is growing even further.

"Observability" is another important area. As AI systems become more complex, the black-box problem of being unable to grasp internal operations intensifies. Observability precisely tracks and visualizes AI's reasoning process and data processing paths, enabling rapid identification of root causes when security incidents occur.

However, applying guardrails excessively degrades AI performance, and false positives in DLP reduce work efficiency. Observability also generates considerable system load if the entire process is to be tracked. CloudNetworks plans to focus on securing the right balance between security, performance, and usability. Through this, we will help enterprises safely adopt AI, reduce the risks of AI use, and generate real business results.

Q5. Do you have any message for employees as we head into 2026?

In 2025, CloudNetworks achieved record-high revenue, laying the groundwork for growth. This was an achievement that all of you created together.

2026 is an important year for solidifying our position as the adoption and use of AI technology accelerates in earnest. The faster the pace of change, the greater the significance of the technical capabilities and experience we possess will become. Based on this, we will help more companies utilize AI safely and efficiently. Your expertise and passion, along with the enjoyment of working together and active communication, are our competitiveness. I hope that in 2026, we will move forward together and that it will be a year in which CloudNetworks leaps forward.

Source: Daily Secu (https://www.dailysecu.com)