Dell Technologies (DELL) The Six Five Summit: AI Unleashed 2026 summary

AI infrastructure evolution for agentic AI

• Agentic AI shifts infrastructure needs from compute-centric training to memory-centric inference, requiring new hardware priorities and a greater role for CPUs in orchestration and coordination across agents.

• Scaling agentic AI from pilots to production introduces challenges in latency, data velocity, and the need for general-purpose, reusable platforms rather than bespoke environments.

Impact on work and organizational structure

• Agentic AI moves beyond unlocking data to digitizing and automating work, fundamentally changing job roles by shifting repetitive, hygiene, and coordination tasks to machines while humans focus on expert and human element work.

• Jobs evolve rather than disappear, as agents handle specific types of work within jobs, requiring organizations to rethink roles and processes rather than simply reducing headcount.

• Organizational transformation demands top-down governance, prioritization, and a willingness to rethink every process and job, moving away from incremental improvements to holistic change.

Economic and infrastructure considerations

• Token economics and ROI for agentic AI require a hybrid approach, matching the cost and capability of AI models to the value of the work performed, with diverse sources of intelligence and infrastructure.

• Hybrid architectures are essential to balance performance, compliance, and cost across a spectrum of AI workloads, avoiding a one-size-fits-all model.