AI Cybersecurity Adoption Surges 56% While Trust Plummets

Organizations are racing to deploy artificial intelligence in their cybersecurity operations, but growing adoption has exposed a troubling pattern: the technology frequently fails at critical moments, and most teams lack the governance frameworks to manage it effectively.

Active AI use in cybersecurity strategy jumped from 50% to 78% among IT and security professionals in just one year, according to the 2026 SANS AI Survey Insights. Yet this rapid expansion has coincided with a sharp increase in reported failures rather than improved security outcomes.

The trust gap widens

Sixty-three percent of practitioners now report significant AI shortcomings in threat detection and response, up from 45% in 2025. Two-thirds said AI guidance has steered them wrong at least once in the past year. Only 27% of practitioners characterized their AI deployment as mature, with most organizations keeping AI in supporting roles or pilot phases.

One anonymous survey respondent captured the prevailing skepticism: "We tend to treat our AI as a digital intern and check its work."

Attackers move faster than defenders

The implementation struggles leave organizations exposed at a dangerous moment. Seventy-eight percent of organizations experienced confirmed or suspected AI-enabled attacks in the past year. Despite 95% of security leaders believing cyber attackers already use AI, only 16% have shifted their defensive strategies to counter AI-driven threats.

The SANS Institute surveyed 536 IT and security practitioners globally, plus 57 senior security executives including chief information security officers.

Governance remains fragmented

A stark disconnect exists between leadership and operational reality. Half of surveyed leaders said their organizations have formal AI risk programs, but only 36% of practitioners reported the same.

"A program that the people doing the work cannot see is not governing much in practice," the survey noted.

Nearly half of respondents reported time and cost savings from reduced manual work. However, organizations rarely track a more critical metric: the share of real threats caught by AI.

"Efficiency is the easiest benefit to see and the easiest to overweight," the survey warned. "An AI system can cut analyst workload sharply while still missing a meaningful fraction of genuine threats, and a team watching only efficiency may not notice until an incident forces the issue."

Why it matters

The widening gap between AI deployment and AI effectiveness in cybersecurity represents a systemic risk for organizations across sectors. As attackers increasingly leverage AI capabilities, defenders who adopt the technology without proper governance or verification processes may create a false sense of security while remaining vulnerable to sophisticated threats. The findings suggest that rapid adoption without mature implementation frameworks could leave organizations worse off than manual approaches.

These findings were first reported by the Journal of Accountancy, based on the 2026 SANS AI Survey Insights from SANS Institute.