Coming soon - Get a detailed view of why an account is flagged as spam!
view details

This post has been de-listed

It is no longer included in search results and normal feeds (front page, hot posts, subreddit posts, etc). It remains visible only via the author's post history.

58
Morgan Stanley Reports Microsoft to Double Capex For AI - NVDA is Clear Beneficiary
Post Flair (click to view more posts with a particular flair)
Post Body

Microsoft (NASDAQ:MSFT) plans to almost double its capital expenditures from $32B in fiscal year 2023 to $63B in fiscal year 2025, according to a forecast by Morgan Stanley.

Much of the new capex is devoted to generative artificial intelligence, but some investors question how much return on investment this spend will produce.

Morgan Stanley expects capex to increase from $16B to $23B for data center capacity to support expanding GenAI revenue streams from Azure AI Services, M365 Copilot and GitHub Copilot.

"This supports a ramp in GenAI revenues from $5.3B in FY24 to $37.9B in FY27, with an implied assumption of GenAI gross margins of ~50% in FY27," said Morgan Stanley analysts Keith Weiss, Josh Baer and Theodor Thun, in a research note on Wednesday.

Microsoft's non-GenAI related data center spend is forecasts to rise from $32B to $48B from fiscal year 2024 to fiscal year 2027.

"This supports a ramp in non-GenAI Cloud revenues from $132B in FY24 to $215B in FY27, with capex intensity modestly improving from 24% to 19% during that period," Weiss noted.

Capex growth outside of data center build outs is much less extensive. Morgan Stanley forecasts this area to grow modestly from $3.7B during fiscal year 2024 to $4.8B by fiscal year 2027.

"While the analysis does suggest Microsoft's GenAI market share exceeds its current software share, our survey work is very supportive of Microsoft accruing additional IT Wallet Share as investment in GenAI ramps," Weiss added. "That said, the limited implied growth in GenAI capex seems at odds with the broader industry discussion on ramping data center build outs and highlights several components necessitating future analysis."

One key variable in the analysis is the expense of expanding computing platforms for OpenAI's training models.

"With little detail on the magnitude of these efforts, we assume figures smaller relative to those speculated within OpenAI," Weiss said

Author
User Disabled
Account Strength
0%
Disabled 2 weeks ago
Account Age
3 years
Verified Email
Yes
Verified Flair
No
Total Karma
383,327
Link Karma
318,428
Comment Karma
64,783
Profile updated: 1 week ago

Subreddit

Post Details

We try to extract some basic information from the post title. This is not always successful or accurate, please use your best judgement and compare these values to the post title and body for confirmation.
Posted
5 months ago