Enterprise AI agent adoption outpaces security controls, report finds

New research from Rubrik Zero Labs indicates that the rapid adoption of AI agents within enterprises is creating a significant security gap, as organizations deploy these autonomous systems without adequate governance and control mechanisms, according to reports by TahawulTech.

The report, based on a survey of over 1,600 IT and security leaders, reveals that 86% anticipate AI agents will surpass their organization's security guardrails within the next year. Compounding this issue, only 23% claim full visibility into the AI agents operating in their environments, leading to an inability to secure identities that are actively making decisions and interacting with critical data. This "shadow workforce" of non-human identities is proliferating faster than enterprises can track, creating new avenues for compromise.

Furthermore, over 80% of respondents find that AI agents require more manual oversight than they provide in efficiency, and 88% lack the ability to roll back agent actions without disrupting systems. With nearly half of respondents expecting agentic systems to drive the majority of attacks in the coming year, the implications for boards and executive teams are clear: AI strategy must be integrated with resilience strategy to maintain operational safety in an increasingly autonomous landscape.

Source: TahawulTech