New report finds 98% of security leaders are deliberately constraining agentic AI deployments due to insufficient access controls and identity management.
Nearly all cybersecurity leaders are actively slowing the adoption of agentic artificial intelligence (AI) systems due to security concerns, according to a new report that highlights a significant gap between AI enthusiasm and practical implementation in enterprise settings.
The 2026 State of Agentic AI Cyber Risk Report, released by cloud-native privilege access management platform Apono, found that 98% of cybersecurity respondents say security and data concerns have already slowed deployments, added review steps, or reduced the scope of agentic AI and autonomous system projects.
The findings suggest that while organizations recognize the potential of AI agents—autonomous systems that can perform tasks and make decisions without human intervention—security readiness is emerging as the primary constraint on widespread adoption.
“Cybersecurity leaders are actively slowing agentic AI adoption,” says Rom Carmel, chief executive officer and co-founder of Apono, in a release. “There’s a lot of talk about AI agents rapidly taking over enterprise workflows, but the data in our report shows that this simply isn’t the case. On the ground, chief information security officers are pressing the brakes.”
Security Gaps Create Implementation Barriers
The report reveals varying degrees of deployment constraints, with 77% of respondents reporting moderate slowdowns or added scrutiny, while 21% cite significant delays or reduced project scope. All respondents agreed that attacks targeting agentic AI workflows would be more damaging than traditional cyberattacks.
Despite widespread interest in AI automation, only 21% of organizations say they feel prepared to manage attacks involving agentic AI or autonomous workflows. Additionally, 98% report friction between accelerating AI adoption and meeting cybersecurity priorities.
“Organizations are still struggling to secure human access at scale,” says Ofir Stein, chief technology officer and co-founder of Apono, in a release. “Expecting chief information security officers to greenlight broad autonomy to agents without mature identity and access controls in place isn’t realistic.”
The study’s results stand in contrast to broader market narratives suggesting rapid, near-term replacement of traditional software by AI agents, according to the report. While experimentation with agentic AI is underway across many organizations, security leaders are implementing additional controls as systems approach production environments.
The report indicates that organizations need stronger controls around identity, access, and permissions before autonomous AI systems can safely scale in enterprise environments.
ID 408706970 | Ai Agent © Gaewchiangmai | Dreamstime.com
