Splunk has released its “State of Security 2025” global research report, highlighting the mounting challenges faced by Security Operations Centers (SOCs). The report uncovers the pain points that mire down organizations and open their doors to threats – 46% of respondents said they spend more time maintaining tools than defending the organization, while only 11% trust AI completely for mission-critical tasks. Furthermore, 66% experienced a data breach in the past year, making it the most common security incident.
With new threats such as AI-powered attacks, organizations must be fully prepared and confident in protecting themselves and their customers. The common thread in addressing these concerns is to build a unified SOC that combines human expertise with AI advancements.
“Organizations are increasingly leaning on AI for threat hunting and detection, and other mission-critical tasks, but we don’t see AI taking complete oversight of the SOC – for good reason,” says Michael Fanning, CISO at Splunk. “Human oversight remains central to effective cybersecurity, and AI is used to enhance human capabilities to help where it truly matters: defending the organization.”
Security teams plagued by technological inefficiencies while external threats increase
When SOC workflows aren’t operating at their peak, it creates major barriers to effective threat detection and response. The report highlights areas of inefficiencies that create risk for organizations:
- 59% say tool maintenance is the main source of inefficiency
- 78% say their security tools are dispersed and disconnected
- 69% say disconnected and dispersed tools creates moderate to significant challenges
Tool maintenance, data silos, and alert fatigue bog down SOC teams. These day-to-day burdens drain valuable time and impact an analyst’s ability to respond quickly and decisively. The report revealed:
- 57% report losing valuable investigation time to data management gaps
- 59% have too many alerts
- 55% have to address too many false positives
SOC analysts are overworked and understaffed
Beyond operational hurdles, the report sheds light on the immense pressure for SOC analysts. High stress levels, chronic understaffing, and burnout are taking a toll and put talent retention and long-term team stability at risk. Findings show that:
- 52% say their team is overworked
- 52% say stress on the job has prompted them to think about leaving cybersecurity altogether
- 43% face unrealistic expectations by leadership
GenAI in the SOC is paying long-term dividends for organizations
Organizations see how AI can alleviate operational and staff shortage problems, as 59% have moderately or significantly boosted their efficiency with AI. Over half (56%) have prioritized the application of AI to security workflows this year, while 1 in 3 (33%) plan to fill skills gaps with AI and automation.
Compared to publicly available tools, 63% agree that domain-specific AI significantly or extremely enhances security operations. However, AI is not running solo as organizations keep humans in the loop to deliver trustworthy AI outcomes. The top three tasks that GenAI is helping across SOCs included:
- Threat intelligence analysis (33%)
- Querying security data (31%)
- Writing/editing security policies (29%)
A unified approach accelerates operations
Minimizing tool maintenance is just the starting point for the benefits of a unified security platform. Adopting a unified approach for threat detection and response leads to tighter collaboration, bringing more context and speed to investigations. Sharing information across security and observability isn’t fully embraced yet, but those who have made the leap report noteworthy advantages. Specifically, 78% of respondents cited faster incident detection, and 66% noted quicker remediation as moderate to transformative benefits.
