Cloud Security
,
Critical Infrastructure Security
,
Fraud Management & Cybercrime
Gigamon 2025 Survey: 17% Increase in Attacks as Public Cloud Vulnerabilities Mount
Artificial intelligence is transforming the enterprise landscape, with organizations reporting a 17% jump in cyber breaches over the past year and 55% experiencing attacks. Security teams struggle with visibility gaps while adversaries weaponize AI to strike harder and faster, according to the Gigamon 2025 Hybrid Cloud Security Survey.
AI is exposing critical security weaknesses faster than most organizations can respond, the survey says. With global AI investment expected to grow beyond $200 billion in 2025 and reach $750 billion by 2028, security leaders must recalibrate how they define, measure and mitigate risk in this new era.
The survey showed that 91% of leaders admitted making compromises in how they secure and manage their hybrid cloud environments. These are not tactical concessions but foundational gaps in visibility.
AI-powered ransomware is surging, with 58% of respondents seeing an increase in the past year alone, compared to 41% in 2024. Nearly half reported attacks targeting their organization’s large language model deployments. Adversarial AI is flooding environments with realistic phishing campaigns, deepfake impersonations and traffic so dense it conceals data exfiltration.
Defenders need to contend not only with rising data volumes but also with the need for clarity. One in three organizations said their network data volumes more than doubled in two years, fueled by AI workloads. Traditional tools are buckling under the strain, and 46% of security and IT leaders now rank managing AI-generated threats as their top security priority when deploying new AI technologies.
Public Cloud Security Concerns Grow
The public cloud once symbolized speed, flexibility and scalability. It now represents unacceptable risk for security leaders.
Public cloud environments are more dangerous than any other infrastructure layer, said 70% of the survey respondents. Over half are actively avoiding AI deployments in public cloud services due to concerns over intellectual property protection and governance gaps. Seventy percent of respondents consider moving data and workloads back to private cloud or on-premises environments.
This shift reflects deeper awareness that agility means nothing without control. As AI expands attack surfaces, security teams are rethinking not only where their data lives but also how visible, governable and defensible that environment is.
Deep Observability Emerges as a Solution
At the core of every compromise is a missing link: visibility.
Nearly half of organizations – 47% – still lack comprehensive insight into their hybrid infrastructure. That includes lateral East-West traffic, encrypted traffic flows and the dynamic data generated by AI workloads. For 64% of respondents, real-time visibility into all data in motion has become their top security priority for the year ahead.
Deep observability provides a solution. Unlike conventional monitoring solutions, deep observability fuses traditional metrics, events, logs, traces data with rich, network-derived telemetry including packets, flows and metadata. This combination provides unmatched clarity into what’s happening across complex, distributed environments.
More than a technical upgrade, deep observability is a strategic necessity:
- 88% of leaders said it’s critical for securing AI deployments;
- 83% said it’s being discussed at the board level.
By closing visibility gaps and delivering actionable insight, deep observability delivers greater control over security.
CISOs Face Mounting Pressure
CISOs stand at the center of this transformation, without the authority or resources they need.
Nearly every CISO surveyed – 97% – acknowledged making security trade-offs, often due to fragmented tooling, shadow AI deployments or lack of influence in key decisions. But 86% said access to packet-level data and application metadata is essential to improving threat detection and response. CISOs aren’t calling for more tools; they’re calling for better integration, cleaner data and visibility that supports strategic action.
Pressure is rising. With 8 of 10 CISOs now believing that cybersecurity carries the same accountability as financial or legal risk, the cost of inaction has never been higher.
Hybrid Cloud Security Standards
The hybrid cloud isn’t going away. Neither is AI. But both demand a fundamental reset in how security is approached, moving from tactical firefighting to proactive control.
Visibility is how CISOs reclaim control, how boards make informed decisions and how organizations turn fragmented signals into a coherent picture of risk. Deep observability doesn’t just meet today’s challenges, it positions organizations to thrive in whatever comes next.
Explore the Gigamon 2025 Hybrid Cloud Security Survey including “Evolving Hybrid Cloud Security in the Age of AI” and “CISO Insights: Recalibrating Risk in the Age of AI” reports that show how security leaders prepare for what’s next.