Organizations are deploying agentic AI at scale despite harbouring fundamental doubts about their underlying data infrastructure. IDC research reveals a stark paradox: whilst 80% of enterprises are funding or running agentic AI applications in production, 45% cite governance and integration concerns as barriers to scaling, and 79% report multiple technical and data-centric challenges. More troubling still, 40% identify data quality as a top concern, yet 75% simultaneously express confidence in their ability to secure their data. This confidence-readiness gap exposes a critical misalignment between executive perception and operational reality. For CX teams already running agents within Zendesk, Salesforce Service Cloud, or similar platforms, this disconnect carries immediate risk: agents trained on poor-quality or poorly governed data will make decisions that damage customer relationships, yet leadership may resist investing in the data infrastructure overhaul required to prevent this.
The bottleneck, as IDC analysts note, is not technological but foundational. Agentic systems are only as reliable as the data they act upon, and the most common use case—agents taking action on live customer data—demands absolute trustworthiness. For support team leads and CX consultants, this means the conversation with stakeholders must shift from "Can we deploy agents?" to "Can we trust our data enough to let agents act autonomously?" Data governance, integration architecture, and quality assurance become non-negotiable prerequisites rather than post-implementation considerations. The risk of overconfidence is material: agents operating on incomplete, siloed, or outdated customer records will escalate tickets incorrectly, miss critical context, or worse, take actions that contradict previous interactions. Organisations that treat data readiness as secondary to agent deployment will find their automation efforts stalling precisely where they promised efficiency gains—in the execution layer where agents interact with live customer systems.
Organizations are confident in their data yet cite governance and security as top barriers to scaling AI agents, say IDC analysts.