Verified, then logged.
Sun 22 Nov 2026 · 00:21 KST / 15:21 UTC한국어 | ENSearch /

08 / Industry · Analysis

Intel sets out chip optimisation strategy for AI workloads

Intel outlined a customer-specific approach linking design and manufacturing around the performance and efficiency requirements of different AI workloads, according to TechM.

A dusty, scuffed, used triple-fan graphics card sits in an open home desktop PC on a wooden desk as out-of-focus hands tighten its bracket screw. AI-generated image

Intel outlined a purpose-built silicon strategy to optimise chips and systems together for artificial intelligence workloads, according to TechM, a Korean technology publication.

According to the report, Intel argued that even small improvements in performance and efficiency become commercially significant as AI deployment expands. That links its customer-specific approach to the demands of operating AI at scale.

TechM reported that Intel proposed connecting design and manufacturing to meet differing customer requirements. The approach starts with the work an AI system needs to perform and extends to how its chips are made.

Workload requirements as the starting point

According to TechM, Marcus Lawrence, who leads custom silicon go-to-market activities in Intel's Central Engineering Group, highlighted the different characteristics of AI workloads in a keynote address. A workload is the computing work a system performs.

The report described Intel's guiding principle as optimising computing performance and efficiency for each workload's requirements. This places the demands of the task at the centre of its design approach.

Connecting customer requirements to manufacturing

According to TechM, Intel's proposed approach connects design, intellectual property (IP), advanced packaging and manufacturing to address customer requirements. Advanced packaging brings semiconductor components together within a package.

TechM also reported that Intel described adjusting chips and systems together, linking efficiency improvements to business competitiveness as AI spreads. The strategy therefore connects workload requirements with decisions across both chip development and the wider system.