Lam Research and Intel surge on BofA upgrades as chip equipment recovery accelerates, SOXX +8.4%

Lam Research jumped 12.7% and Intel gained 9.3% as Bank of America upgrades and improving wafer-fab spending forecasts drove the SOXX semiconductor ETF up 8.4%, its best single-session gain in weeks. Oracle fell 8.5% despite a Q4 earnings beat as investors fixated on mounting capital expenditure commitments.

7 min read
Lam Research and Intel surge on BofA upgrades as chip equipment recovery accelerates, SOXX +8.4%

Lam Research Corp. surged 12.7% to lead a broad recovery in AI chip equipment stocks on Thursday, as a wave of analyst upgrades and improving wafer-fab spending forecasts pushed the SOXX semiconductor ETF to its sharpest single-session gain in recent weeks. The S&P 500 rose 1.75% to 7,394, the Nasdaq Composite added 2.54% to 25,810, and retail-driven enthusiasm surrounding the anticipated SpaceX initial public offering provided an additional lift to AI infrastructure names across the board.

Related startups

Today's biggest movers

Ticker Close Day 1mo YTD
$LRCX $362.52 +12.65% +22.71% +95.89%
$ARM $342.23 +11.32% +54.71% +198.29%
$AMAT $552.64 +11.19% +26.58% +105.54%
$ASML $1,899.48 +9.53% +20.10% +63.22%
$INTC $116.96 +9.27% -2.77% +197.00%
$ORCL $184.10 -8.53% -2.98% -5.93%
$NOW $103.08 -2.81% +18.41% -30.09%
$CRM $166.45 -2.62% +0.37% -34.37%
$MSFT $390.34 -1.77% -3.67% -17.47%
$BIDU $116.11 -1.17% -22.85% -22.75%

Lam Research surges 12.7% as chip equipment spending forecasts signal multi-year AI upcycle

Lam Research Corp. (NASDAQ: LRCX) topped the AI stock leaderboard Thursday with a 12.7% advance to $362.52, its largest single-session gain in recent months. The catalyst was a convergence of analyst target increases and growing conviction around multi-year wafer-fab equipment demand linked to AI data center construction. Cantor Fitzgerald analyst C.J. Muse raised his price target to $425 from $320, citing accelerating AI and DRAM-related capital spending; UBS lifted its target to $375 from $310; and Barclays analyst Tom O'Malley maintained an Equal-Weight rating but raised his target to $335 from $275. With shares now up 95.9% year-to-date, Lam sits among the strongest performers in the chip supply chain. Barclays noted, however, that the current forward price-to-earnings multiple embeds optimistic long-term wafer-fab spending assumptions that could leave the stock exposed if capital spending stalls.

ARM soars 11.3% as AI CPU demand thesis attracts fresh analyst conviction

Arm Holdings plc (NASDAQ: ARM) jumped 11.3% to $342.23, extending a year-to-date advance that now stands at 198%. Bank of America Securities maintained its Neutral rating Thursday but raised its price target to $335 from $245, reflecting an upward revision to long-term AI-driven CPU demand assumptions. Analyst consensus fair value for ARM has risen from $237 to $345 as firms price in the company's growing role in agentic AI processing and edge inference. For the current fiscal quarter, ARM has guided for revenue of $1.01 billion to $1.11 billion, with consensus EBITDA near $540 million and implied margins of approximately 43%. The one-month gain of 54.7% marks a decisive recovery from last week's SOXX -10.4% rout that briefly dragged ARM below $310.

Applied Materials rises 11.2% on Q2 earnings beat and multiple analyst target hikes

Applied Materials Inc. (NASDAQ: AMAT) climbed 11.2% to $552.64, supported by strong second-quarter results reported this week and a series of upward analyst revisions. The company posted Q2 2026 revenue of $7.91 billion with earnings per share of $2.86, both above consensus estimates. Barclays raised its price target to $590 from $500, calling out stronger multi-year wafer-fab equipment forecasts and increased AI-linked DRAM spending. Cantor Fitzgerald lifted its target to $650 from $575 and maintained an Overweight rating. UBS analyst Timothy Arcuri increased his target to $570 from $515 with a Buy rating, citing secular demand from AI data center construction. Applied Materials is now up 105.5% year-to-date, placing it alongside LRCX as a direct beneficiary of the accelerating chip equipment cycle.

Intel gains 9.3% on rare BofA double upgrade; Google TPU foundry order anchors bull case

Intel Corp. (NASDAQ: INTC) added 9.3% to $116.96, extending a year-to-date gain of 197% that has made it one of the most consequential turnaround stories in the 2026 AI cycle. Bank of America analyst Vivek Arya issued a rare double upgrade Thursday, moving Intel to Buy from Underperform and raising his price target to $135 from $96. Arya cited two specific drivers: stronger server CPU demand tied to agentic AI workloads and materially improved foundry visibility following reporting that Google plans to commission Intel to manufacture more than three million tensor processing units in 2028. The Google foundry relationship was first flagged in this column's June 8 recap, when Intel surged 11% on the initial news. At $116.96, Intel still trades 12.5% below its May 2026 high of $129.44, suggesting further room to recover if foundry execution holds.

Oracle falls 8.5% despite earnings beat; capital expenditure scale unsettles investors

Oracle Corp. (NYSE: ORCL) dropped 8.5% to $184.10 on Thursday, a day after reporting fiscal fourth-quarter 2026 results that beat on both revenue and earnings per share. Revenue came in at $19.2 billion, up 21% year over year, with EPS of $2.11 against a consensus estimate of $1.89. Cloud infrastructure revenue surged 93% and remaining performance obligations, a forward-revenue proxy, reached $638 billion, up 363% year over year. Investors looked past those numbers and focused on the implied capital expenditure burden: management confirmed substantial ongoing data center investment to service the contracted backlog, and that scale damped enthusiasm for an otherwise strong print. Oracle maintained its fiscal 2027 revenue target of $90 billion and raised its adjusted EPS outlook to $8.05. A consensus of 43 analysts continues to rate Oracle a Buy, with a 12-month average target of $255, implying approximately 38% upside from Thursday's close. The next Oracle earnings call is scheduled for September 8, 2026.

Notable but quieter

ASML Holding N.V. (NASDAQ: ASML) added 9.5% to $1,899.48, buoyed by renewed commentary around the Terafab semiconductor manufacturing project after Elon Musk made a virtual appearance at ASML's annual technology conference; Bank of America raised its price target to EUR 1,921 and Barclays moved to EUR 1,900. Super Micro Computer Inc. (NASDAQ: SMCI) rebounded 9.2% to $31.97, partially recovering from the 28% collapse on Wednesday after the server maker's $7 billion equity offering drew investor ire; the raise was tied to $39 billion in AI server orders from more than 20 customers but the dilution was severe. Advanced Micro Devices Inc. (NASDAQ: AMD) climbed 8.0% to $488.45, up 118.6% year-to-date, with Bank of America setting a 12-month price target of $560. CrowdStrike Holdings Inc. (NASDAQ: CRWD) rose 6.8% to $691.53 and Palo Alto Networks Inc. (NASDAQ: PANW) gained 6.2% to $279.53, both carried higher in the broad risk-on session.

What to watch next week

Samsung Electronics presents at the Bank of America Korea Conference on June 17, where investors will look for commentary on advanced packaging capacity and high-bandwidth memory pricing. Micron Technology is due to report fiscal third-quarter 2026 results on June 24; the company has guided for revenue of $33.5 billion, plus or minus $750 million, and EPS of $18.90, plus or minus $0.40, a print that would set a new bar for AI memory demand if achieved. NVIDIA Corp. holds its Annual General Meeting on the same day, June 24, with shareholder questions on data center buildout timelines and export license outcomes expected to draw attention. ServiceNow Inc. (NYSE: NOW) and Salesforce Inc. (NYSE: CRM) remain under pressure on weak enterprise software sentiment and are both scheduled to report in late July.

Not investment advice.

© 2026 StartupHub.ai. All rights reserved. Do not enter, scrape, copy, reproduce, or republish this article in whole or in part. Use as input to AI training, fine-tuning, retrieval-augmented generation, or any machine-learning system is prohibited without written license. Substantially-similar derivative works will be pursued to the fullest extent of applicable copyright, database, and computer-misuse laws. See our terms.