Morgan Stanley Backs Micron, Broadcom on Chipflation as AI Capex Keeps Outrunning Supply
Updated
Updated · The Motley Fool · Jul 12
Morgan Stanley Backs Micron, Broadcom on Chipflation as AI Capex Keeps Outrunning Supply
3 articles · Updated · The Motley Fool · Jul 12
Summary
Morgan Stanley said persistent “chipflation” is creating a buy-the-dip setup in AI infrastructure, naming Micron and Broadcom as preferred ways to ride hyperscalers’ spending boom.
Record cloud capex and larger AI models are keeping demand for GPUs, HBM and advanced DRAM above supply, while new foundries take years to build and ramp.
Micron stands to benefit from a structural HBM shortage and long-term supply agreements, which Morgan Stanley sees supporting unit growth and cushioning earnings volatility even if memory prices ease from peaks.
Broadcom offers a parallel bet through networking and custom silicon—its switches, optical links and ASIC work with Alphabet, Apple and Meta help scale GPU clusters while reducing customers’ total cost of ownership.
The bank argues AI infrastructure is shifting from breakneck deployment to a more durable phase focused on utilization, token economics and returns on capital, not a collapse in demand.
As chipmakers race to build new fabs, could an AI efficiency breakthrough create a massive chip glut instead of a shortage?
Beyond higher prices for gadgets, what tangible benefits will AI's expensive chip diet deliver to the average person by 2028?
The AI revolution runs on chips and data centers. What is the hidden environmental cost of this massive infrastructure build-out?
Chipflation and the AI Gold Rush: Market Risks, Supply Chain Strains, and the Future of Semiconductor Investment
Overview
The global economy is experiencing 'chipflation,' where rising semiconductor prices are driven by the explosive growth of AI workloads and the massive build-out of AI infrastructure. This surge in demand, combined with ongoing supply chain challenges, is pushing the cost of essential computing hardware higher. Companies are investing heavily in data centers and AI-enhanced devices, leading to an unprecedented need for processing power—so much so that even older GPUs are fully utilized. As a result, industry leaders like NVIDIA have reported record revenues, highlighting the intense market pressure and rapid expansion fueled by AI's insatiable demand.