ReportGem ReportGem 中文

REAL-TIME GLOBAL RESEARCH

Investment strategy for 2H26: Proposing stock strategies for both risk-on and risk-off markets

Published: 2026-07-07Institution: Goldman SachsPages: 40Original language: EnglishEvidence page: 3

Research evidence excerpt

Investment strategy for 2H26: Proposing stock strategies for both risk-on and risk-off markets

Goldman Sachs Asia Technology

2H26: Three key messages and stock recommendations

In this report, we update our investment strategy, focus areas, and stock

recommendations across the Asia technology team coverage for 2H26.

In our January 12 report, Asia Tech Outlook & Resurgent Japan - Seizing the Global

Growth Opportunity: A broader and deeper AI presence in the Asian supply chain, we

highlighted five themes we believe are under the market’s radar in the AI space:

n Extensive business opportunities driven by soaring power consumption

n Extensive business opportunities driven by faster large-scale data transmission

n Tightening supply/demand for niche components/materials

n Acceleration of physical AI

n AI disruption

While these five focus areas remain unchanged, given the significant share price

fluctuations YTD we now present three perspectives for 2H26 and update our stock

recommendations.

Three perspectives into 2H26

1. We remain bullish on fundamentals/share prices in the Asia AI supply chain

(hardware). The two bottom-up trends we monitor for AI server/data center (DC)

applications driving the cycle—namely (1) signs of semiconductor/component device

supply exceeding demand, and (2) signs of a shift toward lower cost trends over higher

value-added technology (i.e., a slowdown in technological evolution)—have yet to

materialize. Furthermore, beyond server (infrastructure) investment, the continued

penetration of edge devices such as physical AI and edge AI could serve as the next

driver. While share prices in the Asia AI supply chain (hardware) have risen significantly

YTD, we maintain an overall bullish stance for 2H26, as we believe the AI cycle is still

large and long.

The English excerpt is extracted automatically from the cited source page and may contain layout or recognition errors. It is never batch translated.

Open report viewer