Amazon, Microsoft, Alphabet, Oracle, CoreWeave seen as key AI compute beneficiaries after Google Cloud Next

Amazon.com Inc (NASDAQ:AMZN), Microsoft Corp (NASDAQ:MSFT), Alphabet Inc (NASDAQ:GOOG), Oracle Corp (NYSE:ORCL, XETRA:ORC), and CoreWeave (NASDAQ:CRWV) are expected to stay cardinal beneficiaries of accelerating AI compute demand, according to a Jefferies expert enactment pursuing attendance astatine Google Cloud Next.

The analysts wrote successful a enactment that the lawsuit reinforced a wide connection that AI infrastructure request continues to outstrip supply, supporting a constructive presumption connected hyperscalers into upcoming earnings.

“Overall, we proceed to beryllium incrementally much affirmative connected AI request vs our checks successful aboriginal April,” they wrote.

Ahead of archetypal 4th fiscal results, investors are focused connected unreality growth, backlog trends, and immoderate changes to superior expenditure oregon capableness plans. Jefferies said it is incrementally much affirmative connected hyperscalers, with the strongest unreality upside skew seen astatine Google Cloud, followed by AWS and Azure.

Jefferies highlighted elevated expectations for AWS, though the steadfast maintained a precocious condemnation presumption connected Amazon. Microsoft is progressively being assessed connected Microsoft 365 and Copilot traction alternatively than Azure bushed size, portion Alphabet remains affirmative connected trajectory but whitethorn beryllium constrained by valuation.

A large taxable was continued AI compute scarcity, with choky proviso crossed GPU generations including Hopper and Blackwell, alongside constricted spot availability for older chips. Industry feedback suggested a increasing displacement toward committed capableness contracts arsenic hyperscalers fastener successful proviso amid persistent shortages.

Capital expenditure crossed unreality providers is besides expected to stay elevated, with Jefferies estimating astir 64% maturation successful 2025 and 58% successful 2026, driven by ongoing AI infrastructure buildout.

The analysts besides pointed to inefficient endeavor usage of AI infrastructure, with debased GPU utilization contempt rising adoption. At the aforesaid time, request is being fueled by competitory unit and “fear of missing out” connected AI deployment cycles.

Further, they noted that enterprises are progressively seeking exemplary agnostic tooling to negociate rising token costs and optimize workloads crossed models.

“A communal absorption that started to look is that enterprises wanted exemplary agnostic harnesses, arsenic they privation to optimize token depletion and person power implicit what exemplary is used, which is pugnacious if they don't ain the harness,” the analysts wrote.

While frontier models from Anthropic, OpenAI, and Google stay wide used, firms are looking for flexibility successful routing tasks to little outgo alternatives wherever possible.