Morgan Stanley flags a rotation out of semiconductors and into AI hyperscalers as the chip trade cools, warning the shift could introduce broader market volatility. The setup creates a relative-value tension between large-cap semis (NVDA, AMD) and the hyperscaler names that buy their chips.
Morgan Stanley flags a rotation out of semiconductors and into AI hyperscalers as the chip trade cools, warning the shift could introduce broader market volatility.
NVDA and AMD face a potential multi-week headwind as Morgan Stanley argues rotation capital moves from semis into hyperscalers — the question is whether NVDA's exceptional fundamentals limit the downside or whether the crowded trade unwinds further.
A surprise large-scale AI infrastructure announcement or blowout hyperscaler capex guidance could re-ignite demand for NVDA/AMD and collapse the short-semis leg before the rotation completes.
CoverageSource: MarketWatch · Published here TUE, JUL 7 · 5:49 PM ET · 2 outlets in this record · latest listed: Yahoo Finance at 5:49 PM ETHow this is decided →
Morgan Stanley is calling for a comeback rally in AI hyperscalers — think Microsoft, Amazon, Google, and Meta — at the expense of the semiconductor trade, which has run hot and is now showing signs of fatigue. The bank's view is that the cooling chip trade is not a one-session blip but a multi-week rotation, and that the churn could make the broader tape choppy in the transition.
NVDA remains the crown jewel of the chip cycle with $215.9B in revenue, up 65.5% YoY, and a remarkable 71.1% gross margin — but those numbers already reflect a near-perfect demand environment. AMD has grown 34.3% YoY as well, though its 12.5% net margin is far thinner and leaves less cushion if demand expectations soften.
The Morgan Stanley thesis hinges on a simple rotation logic: hyperscalers have underperformed semis on a relative basis during the chip mania, and now that semis are cooling, the capex story flows back to the buyers rather than the suppliers. If hyperscaler earnings guidance holds up, that relative trade has a clear catalyst runway.
The risk to the short-semis / long-hyperscaler framing is that NVDA's fundamentals are genuinely exceptional — a 55.6% net margin at scale is rare — and any renewed AI demand signal (a large model launch, a surprise data center order) could re-ignite the chip trade before the rotation fully plays out. The pair is inherently macro-sensitive; any risk-off move would compress both legs.
Morgan Stanley's rotation call is a relative-value setup: long hyperscalers (MSFT, AMZN, GOOGL) against short semis (NVDA, AMD), where semis have front-run the AI capex story and hyperscalers have lagged. NVDA's 65.5% revenue growth and 71.1% gross margin are already consensus-known, limiting upside surprise, while AMD's thin 12.5% net margin makes it more vulnerable to a sentiment shift. The pair works if the rotation thesis plays out over the next earnings cycle.
The read above, as written. kept as written · closes shown from JUL 8 on
3-6 weeks. Follow to be told when one lands.
Price context does not establish that the story caused the move.
NVDA's 55.6% net margin at $215.9B in revenue is structurally rare and suggests the chip trade's fundamentals have not peaked — any demand re-acceleration from a new model cycle could quickly reverse the cooling sentiment.
NVDA's 65.5% YoY revenue growth is already priced into a crowded consensus long, and Morgan Stanley's rotation call signals that incremental capital at current valuation levels is more likely to flow toward the hyperscaler buyers than the chip suppliers.
Kept as written · your side, if you take one, is graded privately against licensed closes after 10 trading days · nothing here is advice · How the Wire is made →
Shares a name with this story — discovery, not a connection.
This page is kept as it was written on Jul 7. Later coverage joins it only when the company and catalyst evidence match, and what the stock did is shown from licensed end-of-day closes — never re-graded, never backdated. The judgment is yours.