Agentic Systems Outpace Enterprise Safety Controls

A new Deloitte report warns businesses are deploying AI agents faster than safety protocols and safeguards can keep up, noting 23% currently use agents and 74% expect adoption within two years while only 21% have stringent governance. The report highlights security, data-privacy, and accountability gaps, and recommends access controls, observability, human approvals, and standards to make agentic systems auditable and insurable in production.