IT Leaders Prepare for Operating Model Shifts as Agentic AI Advances

There are two primary perspectives on the AI-agent workforce: one viewing the human role as transitional to improve accuracy and build trust, and another expecting AI to augment humans in asynchronous, choreographed operations. [1] Organizations will likely adopt a combination of both agentic and human-augmented approaches, with numerous implementations currently in pilot stages. [1] To capture value from this shift, 75% of IT leaders state their operating models and processes must change within the next 12 to 18 months, according to Deloitte’s 2026 Global Technology Leadership Survey. [1]

Upskilling technology professionals for this transition necessitates behavioral changes rather than just training, according to Doug Vargo, VP of consulting services at CGI. [1] Because AI will handle more routine development tasks, engineers must adopt new workflows centered on framing intent, curating context, and validating AI-generated outputs for accuracy and security. [1]

The transition toward agentic AI represents a fundamental shift in IT operations, moving engineers away from routine development and toward governance and validation roles. As human involvement evolves into a transitional phase to ensure accuracy, the demand for behavioral adaptation will outpace traditional technical training. The high percentage of leaders planning operating model changes within 12 to 18 months indicates an urgent timeline for these workforce transformations. What the sources don't address: what specific upskilling pathways or metrics IT organizations should use to measure successful behavioral changes in their engineering teams.