Optics is the next big AI bottleneck. This company could be an underrated beneficiary.

Dow Jones
5小時前

MW Optics is the next big AI bottleneck. This company could be an underrated beneficiary.

By Britney Nguyen

GlobalFoundries just introduced a new technology to boost the use of co-packaged optics in AI data centers

GlobalFoundries' stock was up 4% on Monday.

Goldman Sachs analysts recently called optical networking the next "buy- out-the-store" artificial-intelligence category that could see extreme demand in the short term. Now, GlobalFoundries is looking to cash in on the rush.

Shares of the chip manufacturer $(GFS)$ were up 4% on Monday following the company's introduction of technology for copackaged optics - an advanced chip-packaging technology to package optical transceivers with chips to reduce power consumption and increase bandwidth.

GlobalFoundries said its new optical module solution, known by the acronym SCALE, improves bandwidth density and the ability to scale systems relative to what traditional copper-based wiring offers. The company has been able to increase data transmission capacity on a single optical fiber.

The demonstration is "a fundamental technology milestone that uniquely positions GF to support the industry's shift to CPO and accelerate the adoption of optical scale-up interconnects," the company said in a statement.

Interest in CPO is growing amid the AI data center build-out, with the Goldman analysts pointing to interconnects, optical components and packaging technologies as emerging supply-chain bottlenecks in a note published last month. Wall Street has been divided over whether optical components will cut into demand for traditional copper, though Nvidia (NVDA) CEO Jensen Huang has recently touted the use of both approaches for the company's scale-up networking technology.

See more: It's been one of Wall Street's most heated AI debates - and it may be totally missing the point

GlobalFoundries added that its platform is the first in the industry to exceed requirements for modern AI data-center optical interconnects set by the Optical Compute Interconnect Multi-Source Agreement, which includes Nvidia, Broadcom $(AVGO)$ and Advanced Micro Devices $(AMD)$.

Cantor Fitzgerald analyst C.J. Muse said he's "starting to see secular drivers emerge in the story" for GlobalFoundries, despite its stock being a "massive underperformer" compared with the broader PHLX Semiconductor Sector SOX since the chip maker's first investor day in August 2022.

Muse pointed to the rise of silicon photonics, satellite communications and physical AI as reasons he's more confident that the company can meaningfully improve its top-line compound annual growth rate over the long haul.

GlobalFoundries' photonic integrated-circuits business looks strong, Muse said, and even if the company is only able to capture a sliver of the overall market for silicon photonics, that could be a "significant" opportunity. Additionally, the company's portfolio of industrial technology is expected to get a boost from edge computing, Muse said.

He upgraded the stock to overweight from neutral Monday, saying that the trends will significantly improve the company's earnings profile.

Muse's $80 price target for GlobalFoundries' stock represents upside of 18.5% from its price of about $67.50 on Monday.

Don't miss: These stocks are popping after Amazon and Anthropic expand their chip partnership

-Britney Nguyen

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

 

(END) Dow Jones Newswires

May 04, 2026 14:21 ET (18:21 GMT)

Copyright (c) 2026 Dow Jones & Company, Inc.

應版權方要求,你需要登入查看該內容

免責聲明:投資有風險,本文並非投資建議,以上內容不應被視為任何金融產品的購買或出售要約、建議或邀請,作者或其他用戶的任何相關討論、評論或帖子也不應被視為此類內容。本文僅供一般參考,不考慮您的個人投資目標、財務狀況或需求。TTM對信息的準確性和完整性不承擔任何責任或保證,投資者應自行研究並在投資前尋求專業建議。

熱議股票

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10