Skate to Where the Workload Is Going Examines the Next Enterprise AI PC Decision

Skate to Where the Workload Is Going is a virtual event taking place on July 29, 2026, exploring enterprise AI PC strategy, AI workload placement, and the refresh-cycle decisions IT leaders are making before AI-native workloads fully arrive. The event page lists a start time of 10:00 AM PT / 1:00 PM ET, with Dokyun Kim, Marika Nox of Qualcomm, and Tom Mainelli of IDC as speakers. The strategic question is whether organizations continue stretching aging hardware or adopt a rolling refresh model that can absorb rising memory costs, OS migrations, security demands, and the shift toward on-device AI.

Why This Event Matters

Most AI conversations start with models. This one starts with the machines expected to run them. The official event description frames the next infrastructure cycle around AI PCs becoming a larger share of enterprise fleets by 2027, while IT leaders contend with memory-driven price pressure and a volatile PC supply environment. That puts refresh decisions in a sharper light because waiting can appear financially prudent while quietly increasing the cost of readiness.

The event's title borrows the familiar strategy of moving before the market fully arrives, but the application is practical rather than motivational. Enterprise buyers are being asked to consider where AI workloads should execute, how much work should happen on-device, and how a dedicated NPU changes the economics of relying solely on CPU, GPU, or cloud capacity. In that context, Snapdragon is more than a sponsor name on the agenda. It is part of the broader conversation about silicon becoming an operating decision.

The Workload Question Behind AI PCs

The event page points to a simple but uncomfortable reality: enterprise AI adoption is moving from pilot projects into procurement plans. If AI PCs become the majority of many enterprise fleets by 2027, organizations making refresh decisions in 2026 are not just buying devices for today's software. They are selecting the architecture that will support future AI models, security policies, collaboration platforms, and productivity workloads.

That is why the CPU, GPU, and NPU distinction matters. A dedicated NPU can change how certain AI workloads are processed locally, affecting latency, power consumption, cloud dependency, and the economics of deploying AI capabilities across an enterprise fleet. The event is positioned around that planning horizon, particularly for IT directors and infrastructure leaders specifying hardware for workloads and AI applications that may not yet be fully defined.

The Speakers and Market Context

The speaker lineup brings together product, market, and device research perspectives. Dokyun Kim is listed as Senior Business Development Analyst at Qualcomm, while Marika Nox is listed as Staff Manager, EMEA Compute Sales and Product Marketing at Qualcomm. Tom Mainelli serves as Group Vice President of Device and Consumer Research at IDC, providing a market-analysis perspective alongside Qualcomm's platform and silicon expertise.

That combination matters because enterprise refresh strategy extends beyond technology alone. It sits at the intersection of procurement timing, workforce needs, security requirements, support lifecycles, and the pace at which AI-capable applications become standard business tools. The more useful conversation is not whether every device needs immediate replacement, but how organizations avoid building a fleet strategy around a workload map that is already becoming outdated.

What Attendees Should Watch

For IT leaders, enterprise architects, infrastructure strategists, and technology decision-makers, the most valuable takeaway may be the discussion around refresh cadence. The event description contrasts steady, rolling refresh cycles with one-time large-scale purchases, arguing that resilience comes from phased modernization rather than treating hardware upgrades as periodic rescue missions. That is a procurement discussion, but it is equally a strategic one.

Attendees should pay close attention to how the speakers connect on-device AI, cloud economics, and long-term platform support. Those considerations will determine whether an AI PC refresh represents little more than a hardware upgrade or a longer-term investment in where enterprise computing is heading. In a market where the next AI workload may arrive before the next budget cycle, organizations that prepare earlier may have greater flexibility when demand accelerates.

The Bigger Industry Shift

Skate to Where the Workload Is Going reflects a broader shift from AI experimentation to AI infrastructure discipline. The first phase of enterprise AI centered on gaining access to models and tools. The next phase focuses on where those tools run, what they cost at scale, and how much flexibility organizations retain as software expectations evolve. That makes the device layer considerably more strategic than it once was.

Ultimately, this event is about judgment under uncertainty. No organization can perfectly predict every AI workload that will matter in 2027 or 2028, but every refresh decision either expands or limits future options. For leaders responsible for enterprise computing environments, the question is no longer whether AI will reshape the fleet. It is whether today's fleet strategy is already moving toward the workloads AI will create.