Amazon.com announced a sweeping price adjustment across its cloud‑computing hardware portfolio, citing an unprecedented shortage of high‑bandwidth memory (HBM) chips that has driven component costs skyward. The price hikes, ranging from 45 to 60 percent depending on the instance type, affect everything from GPU‑accelerated compute nodes to specialized AI inference machines. In a statement, Amazon’s head of infrastructure attributed the spike to a confluence of supply‑chain disruptions, geopolitical tensions, and a surge in demand from AI startups and large enterprises seeking to train generative models. The move puts pressure on the broader cloud market, where rivals such as Microsoft Azure and Google Cloud are already grappling with similar cost escalations, potentially prompting a ripple effect that could raise the overall price of compute services.
Customers are reacting with a mix of concern and strategic recalibration. Enterprises that have built cost‑optimisation models around predictable AWS pricing now face budget overruns, prompting some to explore multi‑cloud strategies or to accelerate the migration to on‑premise AI accelerators. Analysts warn that the price shock could slow the rapid expansion of AI workloads, especially among midsize firms that operate on thin margins. At the same time, the shortage underscores the strategic importance of semiconductor manufacturing capacity, reigniting calls for increased domestic chip production and for cloud providers to diversify their supply chains to mitigate future volatility.
This story was written by a language model from reporting by techcrunch.com. Check the original before quoting it.