Big Tech Weekly | Micron Confirms AI Is Lifting Semi; JPM Says AI Valuations Remain Conservative

This Week’s Macro Theme

November employment data showed the U.S. unemployment rate unexpectedly rising to 4.6%. While still low by historical standards, this marks the highest level since early 2021. Under the current macro pricing framework, a moderate softening in labor data is paradoxically viewed as positive by markets.

The Bank of Japan raised its policy rate from 0.5% to 0.75% this week, in line with market expectations. This move lifts Japanese interest rates to their highest level in 30 years and marks the first rate hike in 11 months since January 2025, signaling continued progress toward policy normalization. While the pace remains gradual, the shift has created marginal disruptions to global capital flows and carry trades.

U.S. stocks sold off sharply on Wednesday, but rebounded on Friday following Micron’s earnings beat and the smooth implementation of the BOJ rate hike. Within the MAG7, only $Apple(AAPL)$ and $Microsoft(MSFT)$ ended the week lower, while $Tesla Motors(TSLA)$ and $NVIDIA(NVDA)$ rose 4.8% and 3.4%, respectively.

Big Tech Core View — Micron’s Earnings Confirm AI Is Systemically Lifting the Entire Semiconductor Stack

Market divergence around the AI trade has intensified. Supported by $Micron Technology(MU)$’s strong results, NVDA rebounded sharply, rising 3.41%.

Recent skepticism toward NVDA has centered on three key questions:

  1. Will AI CapEx slow after 2026?

  2. Will TPUs erode GPUs’ competitive moat?

  3. Have gross margins already peaked?

Rather than debating GPU substitution, the more relevant question is the true impact of AI on supply–demand dynamics across the entire value chain.

Bernstein’s latest report argues that these debates are somewhat overcomplicated:

The real question is not TPU vs. GPU, but whether the AI opportunity is large enough.

If the opportunity is large, GPUs will continue to sell, TPUs will continue to scale, and custom silicon, networking, and memory will all benefit together. If it is not, everyone loses.

JPMorgan stated this week that market pricing of AI companies remains too conservative.

Following the recent pullback, AI stocks under its coverage trade at only about a 26% historical valuation premium, which JPMorgan views as clearly insufficient.

AI-driven earnings acceleration should be closer to 60%–80%, yet investors are currently pricing in only ~30% sustainable CapEx growth, versus roughly 70% in 2024–2025.

Micron’s Results Show AI Excess Returns Are Spreading from a Single Processor to the System Level

Micron’s guidance came in well above expectations:

  • FY2Q EPS guidance: ~$8.4, about 75% above consensus

  • Gross margin guidance: 68%+, entering a historically high range

  • FY26 net CapEx: maintained at $20bn, well below market fears of aggressive supply expansion

Morgan Stanley described this as the fastest earnings inflection point it has seen in 32 years of covering the memory industry and reiterated Micron as a Top Pick in U.S. semiconductors.

Unlike the market’s prior focus on HBM, several institutions (MS / Barclays / GS) highlighted a more important signal: systemic shortages are spreading from HBM to DDR5.

Morgan Stanley noted that current memory tightness is no longer confined to HBM, but reflects broad-based DDR5 shortages—evidence that AI is compressing the entire semiconductor supply–demand curve, not just one niche.

Contrary to market concerns, Micron’s CapEx guidance actually reinforces AI’s pricing power. FY26 net CapEx of $20bn is below some aggressive expectations.

Goldman Sachs estimates Micron’s 2026 bit growth at ~20%, while demand shows no corresponding slowdown, implying supply–demand tightness lasting at least through 2026. This closely mirrors NVDA’s environment: expanding demand, restrained supply, and structurally higher through-cycle profitability.

Bernstein’s Core Recommendation: Hold Both NVDA and AVGO

As concerns around AI sustainability have grown, this trade has cooled. However, the companies actually spending money on AI show no signs of slowing. Demand remains strong, and earnings expectations for NVDA and $Broadcom(AVGO)$ still appear understated.

# Nvidia Still A Top 2026 Chip Pick: Already Hit Bottom?

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Report

Comment

  • Top
  • Latest
empty
No comments yet