91% of Security Leaders Are Hitting the Cloud Panic Button

The public cloud honeymoon is officially over. While tech evangelists spent the better part of a decade preaching the gospel of cloud-first everything, a sobering reality check has arrived, according to Gigamon’s latest security survey.

The numbers tell a story that would make any CISO break out in a cold sweat: breach rates have exploded to 55% in the past year alone. That is a 17% jump that coincides suspiciously with AI’s meteoric rise.

“Security teams are struggling to keep pace with the speed of AI adoption and the growing complexity and vulnerability of public cloud environments,” says David Land, Gigamon’s vice president for APAC.

The survey of over 1,000 global security and IT leaders across six countries reveals a perfect digital storm brewing. AI workloads have more than doubled network data volumes for one in three organizations in just two years. 47% are watching helplessly as attackers set their crosshairs on LLM deployments, while AI-powered ransomware attacks have surged from 41% to 58% of respondents in a single year.

The really interesting part is that the very technology that was supposed to democratize innovation is now forcing a mass exodus from the public cloud. In Singapore — often considered a bellwether for regional tech adoption — 71% of security leaders now view public cloud as riskier than any other environment.

The numbers get more dramatic when you dig deeper. A staggering 76% of Singaporean organizations are actively considering pulling their data back from public to private clouds. More than half (54%) are too spooked to even deploy AI in public cloud environments, citing fears about intellectual property theft.

CISOs are in panic mode. The survey notes that 55% of respondents lack confidence in their current tools’ ability to detect breaches. When your security stack has the detection capabilities of a broken smoke alarm, every shadow looks like a threat.

But there may be hope, that is if you believe in “deep observability” — the latest buzzword that’s backing up its hype with substance. Unlike traditional monitoring that skims the surface, deep observability combines MELT data (metrics, events, logs, and traces) with network-derived intelligence like packets, flows, and metadata.

The urgency for deep observability capabilities is palpable. With cybercrime costs ballooning to USD3 trillion globally according to the World Economic Forum, companies can’t afford to play whack-a-mole with threats anymore. The survey shows 64% of leaders are prioritizing real-time threat monitoring over the next 12 months, while 89% consider deep observability fundamental to securing hybrid cloud infrastructure.

Perhaps most telling is that 88% of Singapore respondents say deep observability discussions have reached the boardroom. When C-suite executives who typically view IT as a cost center start demanding granular visibility reports, you know the threat landscape has fundamentally shifted.

Bottom line

For CISOs, this survey represents a watershed moment. The era of “move fast and break things” cloud adoption is over. The new mandate is “move carefully and see everything.” Chief data officers face an equally stark reality — their data gravity is pulling them back toward private infrastructure.

The AI revolution promised to make everything smarter, faster, and more efficient. Instead, it’s making everything more complex, vulnerable, and expensive to secure. Strangely, the future looks suspiciously like a more sophisticated version of the past.

Image credit: iStockphoto/cyano66