· Event impact
Goldman Sachs Sees ASICs Matching GPU Demand by 2027, Threatening GPU Dominance
Transmission path
A projected shift in AI hardware demand from general-purpose GPUs to specialized ASICs creates a new competitive dynamic, potentially diverting revenue and valuation from the current market leader to established ASIC designers.
Market mechanism
A projected shift in AI hardware demand from general-purpose GPUs to specialized ASICs creates a new competitive dynamic, potentially diverting revenue and valuation from the current market leader to established ASIC designers.
Extended read
A new forecast from Goldman Sachs suggests a significant evolution in the AI hardware landscape is on the horizon. The bank predicts that demand for custom-built AI chips, known as Application-Specific Integrated Circuits (ASICs), will grow to match the demand for Graphics Processing Units (GPUs) by 2027. This trend is supported by a Bloomberg Intelligence forecast for 27% annual growth in the ASIC market through 2033. This potential shift has major implications for the semiconductor industry. While Nvidia has dominated the AI training market with its GPUs, the rise of ASICs for AI inference and specialized tasks could redistribute market share. The report identifies Broadcom (AVGO) and Marvell Technology (MRVL) as the primary players poised to capture this growth. Broadcom, in particular, is expected to leverage its strong design capabilities and customer relationships to maintain a commanding 60% share of the custom chip market. This development suggests that the next phase of the AI hardware buildout may be less concentrated in a single player and more distributed among key enablers of custom silicon.
Exposed assets
AVGO · MRVL · NVDA
Continuous event tracking, options routing, and portfolio overlap for this event and the assumptions it moves live inside Market Ontology. Start a trial →