Artificial intelligence (AI) is becoming important to cybersecurity strategies across Asia Pacific (APAC), but many organizations still face practical challenges when trying to integrate the technology into their Security Operations Centers (SOCs), according to cybersecurity company Kaspersky.
A global study conducted by the company found that 99% of respondents from the APAC region plan to incorporate AI into their security operations. Of that number, 67% said they will probably implement AI tools in their SOCs, while 32% said they will definitely do so. The findings show a growing demand for technologies that can strengthen threat detection, speed up investigations, and improve overall SOC efficiency.
Despite the strong interest, the report highlights a significant gap between expectations and real-world implementation.
“Across APAC, organizations are taking a pragmatic approach to AI in the SOC, prioritizing use cases that deliver immediate operational impact,” said Adrian Hia, managing director for Asia Pacific at Kaspersky. “The strongest expectations center on enhancing threat detection through automated anomaly analysis and accelerating response through predefined automation.”
According to the study, organizations in the region primarily expect AI to improve cybersecurity through automated analysis of large data sets to detect unusual behavior or suspicious activity. About 60% of respondents said they plan to use AI to strengthen threat detection, while 55% expect the technology to automate incident response by triggering predefined actions during cyberattacks.
Companies also see AI as a way to ease pressure on cybersecurity teams. The top motivations for adopting AI in SOCs include improving overall threat detection effectiveness (55%), automating routine security tasks (47%), and increasing accuracy while reducing false positives (45%).
These priorities show the region’s focus on improving operational resilience and reducing “alert fatigue,” a common challenge for security analysts who must process large volumes of alerts generated by monitoring systems.
However, turning AI ambitions into operational systems remains difficult for many organizations.
The biggest barrier cited in the report is the lack of high-quality training data, identified by 44% of respondents as a key obstacle. Without reliable and diverse data sets, AI models struggle to accurately identify threats or generate meaningful insights.
Other major challenges include a shortage of AI-skilled cybersecurity professionals (37%), emerging security risks linked to AI technologies (34%), integration difficulties with existing security tools (34%), and the high costs associated with developing and maintaining AI-powered solutions (33%).
Hia said these issues highlight the need for a more structured approach to deploying AI in cybersecurity environments.
He noted that while large enterprises globally are exploring broader applications of AI across SOC functions, many organizations in Asia Pacific remain focused on practical implementations that directly improve day-to-day security operations and strengthen cyber resilience.