Nvidia, Korean AI-chip startup Rebellions and SK Telecom announced a collaboration on Monday, November 9, at Nvidia AI Day Seoul 2026 to advance Korea's sovereign AI infrastructure. Rebellions will connect its next-generation Rebel R200 neural processing units to Nvidia's platform through NVLink Fusion and build rack-scale inference systems on Nvidia's MGX rack architecture; SK Telecom plans to deploy those systems in its AI data centers, linking them to its proprietary A.X K2 foundation model and public-facing services under Korea's government-backed 'AI for All' project. The effort is supported by Korea's Ministry of Science and ICT, which is also expanding broader cooperation with Nvidia covering AI semiconductors, data-center infrastructure, models, agents, talent development and security, including work with Samsung, Kakao Mobility, LG Electronics and Hyundai Motor Group. The partners plan disaggregated inference — Nvidia GPUs for prefill, Rebellions NPUs for decode — to improve throughput and power efficiency. R200 systems are expected to be ready for deployment in 2028; no performance specs, volumes, prices or financial terms were disclosed.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nvidia (NVDA) — extends its NVLink Fusion ecosystem to Korean-designed silicon and deepens its sovereign-AI footprint; strategically positive, though shares fell about 5-6% Monday on the separate board probe into CEO Jensen Huang, which dominated trading. SK Telecom — will deploy the joint systems in its AI data centers, linking them to its proprietary A.X K2 model and the government's 'AI for All' public services. Rebellions (private) — its next-generation R200 NPU gets validated inside Nvidia's platform and MGX rack ecosystem, a major endorsement for the Korean AI-chip startup. Samsung Electronics / SK Hynix — both back Rebellions and supply Korea's AI hardware ecosystem; the deal reinforces Korea's domestic AI-chip ambitions. Alphabet (GOOGL) — its TPU push, announced the same day, is the competing in-house-silicon story.
Google announced two new TPU systems on Monday, November 9, completing its eighth-generation lineup at four systems in a single year — a first in more than a decade of TPU development. TPU 8tx is built for secure and confidential inference (plus small-scale training) for regulated enterprises and governments: it uses the same silicon as the training-focused TPU 8t, pairs it with 5th Generation AMD EPYC host CPUs and the low-latency Boardfly network, building encryption into the hardware layer so data and model weights stay protected even while in active use, with up to 2x lower latency than TPU 8t. TPU 8i Max targets the memory wall for massive context windows: it combines two TPU 8i chips in a single package, doubling the compute, on-chip SRAM and HBM bandwidth of TPU 8i and delivering the highest compute density per rack. Google also opened up TPU development to PyTorch developers, launched its TPU agent suite and XLA Serving Engine, and said shipments of the two new systems are expected next year.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Alphabet (GOOGL) — Google Cloud's in-house AI silicon story strengthens; four TPU systems in one year, first time ever, with nearly 100 customers and all available capacity booked. Nvidia (NVDA) — TPUs are the most credible in-house alternative to its GPUs; industry research cited in coverage forecasts TPU-related sales growing from about $12B this year to more than $135B by 2028, with TPUs costing roughly 30% less than Nvidia GPUs. Advanced Micro Devices (AMD) — its 5th Generation EPYC processors are the host CPUs for the new TPU 8tx system. Broadcom (AVGO) — custom-silicon peer; the TPU ramp validates the custom-AI-chip market but also sharpens competition for hyperscaler accelerator budgets.
Microsoft is winding down a 2023 arrangement to rent AI computing capacity from Oracle, Bloomberg reported on Friday, November 6, citing people familiar with the matter. Microsoft decided not to renew the contract and has spent recent months switching off the rented capacity, which was originally used to power the AI-assisted version of Bing, because the Oracle servers run an older generation of Nvidia chips that no longer meet Microsoft's evolving needs. The full size of the deal was never disclosed and could not be determined. Oracle responded that demand for its AI infrastructure remains strong, including for earlier-generation GPUs, saying GPU capacity that came up for renewal in its fiscal first quarter was renewed or resold at about a 20% premium to prior long-term pricing. Bloomberg framed the move as part of a wider Microsoft reassessment of the more than $80 billion it has committed to renting capacity from outside cloud providers, including CoreWeave, Nscale and Nebius.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Oracle (ORCL) — directly losing a tenant for older-GPU capacity; shares fell to a 52-week low of $109.21 intraday Friday, down 5.2% from the November 5 close, before recovering to close at $116.10, up 0.82%. Microsoft (MSFT) — tightening oversight of outside capacity and reassessing $80B+ in commitments; shares little changed on the report. Neocloud/AI infrastructure names (CoreWeave, Nscale (private), Nebius, IREN) — Microsoft's review puts their contracted revenue under scrutiny
IREN fell nearly 17% on Nov 5 after disclosing Microsoft deployment delays. Nvidia (NVDA) — the rented Oracle servers ran older-generation Nvidia chips, underscoring how fast prior-gen hardware loses hyperscaler demand.
Google has taken over Crusoe's unfinished data-center project in southeastern Wyoming, Crusoe confirmed on Friday, November 6. Google, which Crusoe had tried and failed to land as a customer for the project, is taking over development after reportedly raising concerns about its cost and construction timeline under Crusoe's management. A day earlier, Google announced plans to invest $17 billion to develop the campus, its first data center in Wyoming, powered mainly by a planned 702 MW fuel-cell plant built with Bloom Energy units and contracted through American Electric Power, alongside grid power, gas generation and carbon capture. Neither company disclosed what Google paid Crusoe for the site. Crusoe is still developing another Google data-center campus near Amarillo, Texas.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Alphabet (GOOGL) — committing $17 billion to its first Wyoming data-center campus, a major owned-infrastructure bet that contrasts with rivals pulling back on leased capacity. Crusoe (private, valued at $30.9B) — exits a troubled project but keeps developing a Google campus near Amarillo, Texas. Bloom Energy (BE) and American Electric Power (AEP) — the campus power plan includes a 702 MW fuel-cell development using Bloom Energy units contracted through AEP. Tallgrass Energy / Black Hills Energy — partners on power and gas for the campus. Neocloud/data-center developers — the sale highlights how hard it is for independent developers to deliver massive projects on time and on budget, reinforcing the post-midterm stress on the sector.
President Donald Trump told CNN on Sunday, November 8, that it is 'highly unlikely' he will need to order the US back to war with Iran as long as there is progress in negotiations over Tehran's nuclear program and reopening the Strait of Hormuz, repeating the assertion five times during a 20-minute phone interview with CNN's Jim Sciutto. 'It depends on whether or not they sign a very strong agreement,' Trump said. 'The answer is highly unlikely it's going to be necessary.' He added that he was 'not looking to' escalate the war in a separate CBS '60 Minutes' appearance. At the same time, Trump stressed the military option remains on the table, saying three US aircraft carriers are in the region, 'fully locked and loaded,' and that the US has identified 38 bridges and other infrastructure targets inside Iran it could destroy 'in literally one hour.' He also claimed Iran's air defenses have been destroyed and that Tehran retains only about 10% of its missile capacity and 15% of its drone capabilities.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Taiwan's government moved over the weekend to arrest the rout in its stock market, and the measures took effect on Monday, November 9. The Financial Supervisory Commission cut the cap on intraday short-sale orders for borrowed shares to 3% of a stock's 30-day average daily volume, down from 30%, and raised the minimum margin for short sales to 130% from 90%, while encouraging buybacks. At an emergency meeting on Friday afternoon, the National Financial Stabilization Fund authorized its tenth-ever market-rescue operation, putting NT$500 billion (about $15 billion) on standby to buy stocks. The Taiex surged as much as 6.7% intraday and stood up 5.79% at 42,001.22 in midday trading, its biggest one-day gain since July 31, ending a seven-session losing streak; TSMC gained about 4.7%, MediaTek more than 4% and Foxconn about 6.5%. South Korea acted in parallel, suspending the margin-loan collateral rule that forces brokers to liquidate leveraged retail accounts; SK Hynix gained about 4.6% and Samsung about 2.4%. Nasdaq 100 futures, down about 0.8% at Sunday's open on the Nvidia probe news, recovered to roughly flat by Monday morning.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
TSMC (TSM): gained about 4.7% in Taipei; the world's largest contract chipmaker and Nvidia/Apple supplier is the direct beneficiary of the stabilization measures and the halt in forced selling
Nvidia (NVDA): the AI bellwether whose shares were indicated about 3.4% lower in overnight trading on the board probe; a stabilized Asian chip tape removes one leg of the selloff pressure, though the probe overhang remains
Memory chipmakers (Micron MU, SK Hynix, Samsung, SanDisk): SK Hynix gained about 4.6% and Samsung about 2.4% as investors judged the memory selloff excessive; Goldman Sachs' AI memory basket had fallen 51% from its June peak
Broadcom (AVGO), AMD (AMD), and the Philadelphia Semiconductor Index (SOX): U.S. chip names that led Friday's partial rebound and track the same AI-infrastructure sentiment the intervention is trying to steady
Foxconn (Hon Hai): gained about 6.5% in Taipei alongside the broader Taiwan rebound
China's State Council Tariff Commission announced on Monday, November 9, 2026 that it will maintain the suspension of an additional 24% tariff on US imports for two more months, effective 13:01 on Tuesday, November 10, while retaining a 10% additional tariff. The move was announced by China's finance ministry on its website and reported by Reuters at 02:48 UTC. The extension implements the consensus reached by Presidents Xi Jinping and Donald Trump at their September summit in Washington. Beijing also said it will closely monitor potential US moves to replace tariffs imposed under the International Emergency Economic Powers Act (IEEPA) — which courts have held invalid — with Section 301 tariffs, and adjust its retaliatory measures as appropriate.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nvidia's board said Sunday, Nov. 8, that it has launched an independent investigation into reports that co-founder and CEO Jensen Huang was the target of a foreign intelligence effort. A committee of independent directors, assisted by outside independent counsel, will oversee the inquiry. In a message to employees on Sunday, Huang acknowledged he had been unfaithful to his wife and apologized to his staff, while maintaining that no proprietary information was shared. The board said it has 'no reason to believe that either national security or NVIDIA secrets were compromised' and expressed confidence in Huang, who supports the investigation and has agreed to cooperate. The reports at issue say U.S. officials concluded Alice Huang, an executive at Singapore-based Megaspeed, was recruited by China's Ministry of State Security to gather intelligence on Huang and potentially influence his stance on U.S. chip-export restrictions. Nasdaq-100 futures were indicated modestly lower after the news broke Sunday evening.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
NVDA: the world's most valuable chipmaker; CEO at the center of a national-security probe, with Nasdaq-100 futures indicated modestly lower on the news
Semiconductor/AI peers (AMD, Broadcom, Micron, TSMC): any spillover from a U.S.-China chip-policy escalation or renewed scrutiny of AI-chip sales to Asia
Megaspeed-linked supply chain: the Singapore firm is itself under U.S. investigation over restricted Nvidia chip access
SPY/QQQ: Nvidia carries a record ~8% weight in the S&P 500, so a company-specific shock mechanically moves the broad market
Berkshire Hathaway reported third-quarter 2026 results on Saturday, November 7. Operating earnings fell 5.8% to $12.705 billion ($8,954 per Class A share, roughly in line with the $8,822 FactSet consensus) from $13.485 billion a year earlier, while revenue rose 11% to $105.65 billion, above estimates of about $101 billion. Net earnings, which include swings in the stock portfolio, fell 28% to $22.11 billion. The headline move was capital allocation: Berkshire bought back about $9.6 billion of its own stock in the quarter, more than double the $4.5-4.8 billion spent in the first half of the year and a quarterly record topping the $9.3 billion set in Q3 2020, plus roughly another $2 billion in the first four weeks of October. Insurance drove the earnings weakness: underwriting profit fell 43% to $1.352 billion and insurance investment income fell 14%, with GEICO's pretax profit down about 45% on higher accident losses, more advertising and price cuts.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Berkshire Hathaway (BRK.A/BRKB) — record buyback and Abel's capital-allocation signal are directly supportive, though the insurance earnings miss tempers it; the stock was already at a record high. Insurance sector (GEICO is a major auto insurer) — the 43% underwriting-profit drop and 45% GEICO decline flag rising accident losses and pricing pressure across auto insurance. Apple (AAPL), Coca-Cola (KO), Chevron (CVX) and other large Berkshire holdings — the 10-Q due later in November will show which stocks Berkshire traded in Q3
Berkshire was a net seller of about $400 million of equities in the quarter after buying roughly $20 billion in Q2.
US forces carried out limited airstrikes against an Iranian air-defense system on Qeshm Island in the Strait of Hormuz on Friday night, November 6. President Trump confirmed the action to reporters after landing in West Palm Beach, Florida, calling it a 'minor' strike: 'They built a little bit of an anti-aircraft device and we knocked it out in about 15 minutes, maybe five.' Iranian state media reported two missiles hitting the Suza area of Qeshm Island, without identifying targets or damage. Iran's army said early Saturday it shot down a US MQ-9 drone over the island, a claim the US has not confirmed. Trump insisted 'we are not fighting' and said the US has made 'a lot of progress' in negotiations with Iran, mediated by Qatar.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Energy/oil majors and refiners (XOM, CVX) — any Hormuz disruption lifts crude and energy equities; defense contractors (LMT, RTX, NOC) — sustained regional conflict supports orders; airlines and shipping (UAL, DAL, AAL) — higher fuel costs and rerouting risk; broad market (SPX, QQQ) — escalation risk is a headwind, though Friday's session showed markets treating this as contained; gold and Treasuries as safe havens if escalation follows.
The Washington Post reported Friday, November 6, 2026 that Democrats are preparing a series of congressional investigations into major U.S. companies — including Apple, Meta, and businesses linked to Elon Musk — after winning majorities in both the House and Senate in the midterm elections. Planned lines of inquiry include Apple's hosting of apps that track U.S. Immigration and Customs Enforcement activity, Meta's demands that local governments sign nondisclosure agreements related to data-center development, and President Trump's reported use of Musk's Grok AI chatbot in planning the military operation that captured Venezuelan President Nicolas Maduro. Separately, Senator Richard Blumenthal (D-Conn.), set to chair the Senate Permanent Subcommittee on Investigations from January 2027, called on Apple, Meta, Google, Nvidia, Amazon, Microsoft and other donors to President Trump's White House ballroom project to voluntarily turn over related records or face possible subpoenas in January. Apple and Nvidia each pledged $25 million toward the ballroom, among the largest donations in a fundraising effort that raised roughly $333 million. Companies and executives are already seeking legal advice on responding to subpoenas, hearings and information requests, and Washington law firms are pitching Musk on personal counsel ahead of an expected barrage of subpoenas.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
AAPL — named early target over hosting of ICE-tracking apps and its $25M ballroom donation; document demands with subpoena threat in January
META — named early target over nondisclosure agreements demanded from local governments on data-center projects; also a ballroom donor facing records demands
TSLA / SpaceX (Musk-linked) — Trump's reported use of Musk's Grok chatbot in planning the operation that captured Venezuela's Nicolas Maduro is a planned line of inquiry; Musk expected to face a 'barrage of subpoenas'
NVDA — named among ballroom donors called on to turn over records ($25M pledge); faces same subpoena threat
AMZN, MSFT, GOOGL — also named in Blumenthal's records demands over ballroom donations; broader Mag 7 regulatory overhang
Broader mega-cap tech — investigations are expected to move fast with subpoenas and public hearings, keeping the whole group under political scrutiny into 2027
Tesla CEO Elon Musk announced on X early Friday, November 6, that Tesla will wind down Cybertruck production at Gigafactory Texas in Austin around mid-2027 to make room for increased production of the Cybercab, Tesla's fully autonomous two-seat robotaxi. Musk did not say whether Cybertruck production would move to another plant or end entirely; Gigafactory Texas is the only known site equipped to build the truck's stainless-steel body. Cybertruck sales have run at roughly a tenth of plan — about 16,000 sold in the first nine months of 2025 (Kelley Blue Book), down 38% year over year, against a line built for 250,000 units a year and installed capacity still listed above 125,000. Cybertruck lead engineer Wes Morrill later indicated existing trucks will keep parts and service support.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Tesla (TSLA) — Cybertruck is a small revenue contributor but the move is the clearest signal yet of the autonomy-first pivot; Cybercab capacity and robotaxi timelines are now the operative story
EV / auto sector — confirms weak demand for premium electric pickups and Tesla's shift of capital from vehicles to robotaxis and robotics
U.S. employers added 165,000 jobs in October, far above the roughly 59,000-65,000 economists had forecast, and the unemployment rate ticked down to 4.1% from 4.2%, the Bureau of Labor Statistics reported Friday at 8:30 a.m. ET. The gain was more than triple the average monthly increase of 48,000 over the prior 12 months. The BLS also revised prior months sharply higher: August's gain was raised by 53,000 to 186,000 and September's by 31,000 to 60,000, a combined 84,000 more jobs than previously reported. Average hourly earnings rose 0.4% to $38.00 in October and are up 3.1% over the past year, while the labor force participation rate held at 61.9%.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Broad market (S&P 500, Nasdaq, Dow): a much stronger-than-expected labor market gives the Fed more scope to focus on inflation, reinforcing December hike expectations. Rate-sensitive growth and AI/tech stocks: a hotter jobs print complicates the post-selloff bounce, since it argues against near-term easing. Banks and financials: higher-for-longer rate expectations are generally supportive of net interest margins. Small caps (Russell 2000): more exposed to higher borrowing costs.
President Trump posted on Truth Social on Friday morning, November 6, reiterating his support for artificial intelligence as the market rout deepened, writing: 'As President, I am 100% in favour of Super Intelligence (SI), and will work with the many States in the US that want to become RICH by having them. They are GOLD, and only foolish leadership will say "NO" to the riches they bring!' In a separate post he blamed voters for the market chaos, writing 'You only have yourselves to blame!' The posts came a day after the Nasdaq's 5.52% plunge and as Nasdaq-100 futures staged a roughly 1.8% premarket bounce ahead of the 8:30 a.m. ET October jobs report.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
AI/data-center exposed names at the center of the rout — Nvidia (NVDA, -9.3% Thursday), Microsoft (MSFT, declined to double its Nscale deal), Meta (META), Alphabet (GOOGL), Amazon (AMZN), Tesla (TSLA, -10.8% Thursday) — plus data-center operators and AI infrastructure names (CoreWeave, IREN, Applied Digital, Nebius) that fell 12%+ Thursday. A presidential pushback against the data-center crackdown narrative is a modest tailwind for all of them, though it does not change the new Congress's composition.
OpenAI president and co-founder Greg Brockman said on Wednesday at the Wall Street Journal Tech Live conference in Napa, California, that OpenAI could be ready for an initial public offering in 2027. Asked directly whether the company would be ready to go public next year, he replied, 'I think so. Yes.' Brockman also said enterprise customers, once about 30% of revenue, now make up the majority of OpenAI's sales, and that the first half of 2026 had been difficult as Anthropic gained ground in enterprise AI and coding tools. OpenAI confidentially filed a draft S-1 in June 2026; no public filing, date, or valuation has been confirmed.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nasdaq futures rose about 1.8% in Friday premarket trading, clawing back part of Thursday's historic selloff, while S&P 500 futures gained roughly 0.6% and Dow futures slipped about 0.55%. AI stocks that had plummeted Thursday rallied in the premarket, with Nvidia up about 4.1%. The bounce came after the AI selloff spread across Asia overnight: South Korea's Kospi closed more than 6% lower in its worst day since 2025 after a 20-minute trading halt, Japan's Nikkei 225 neared correction territory, and indexes in Hong Kong and Shanghai also fell. The October nonfarm payrolls report is due at 8:30 a.m. ET, with consensus around 59,000-65,000 new jobs and unemployment steady at 4.2%.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nvidia (NVDA) — up ~4.1% premarket after losing $374-447 billion in value Thursday
Tesla (TSLA) — fell ~9.5-12% Thursday, also bouncing in premarket; broad AI/semiconductor complex (Broadcom, AMD, Micron) — leading the premarket recovery after the SOX's 12.25% Thursday plunge
Asian tech (Samsung, SK Hynix, TSMC-linked names) — still falling overnight as the rout spread to Kospi and Nikkei.
Microsoft is declining the option to double the size of its deal with Nscale to rent more computing power from a Texas data center, the Financial Times first reported on Nov 5; separately, Bloomberg reported Microsoft is tightening oversight of its data-center projects and stepping away from projects it deems risky. Internally, Microsoft is reconsidering advancing deals with its data-center partners, per Bloomberg. A Microsoft spokesperson told Reuters the company's approach to infrastructure 'is built on flexibility and optionality, combining Microsoft-owned data centers, leased facilities and strategic partnerships.' Microsoft shares fell 3.7% on Thursday.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Micron Technology has all but halted construction on its planned Fab 11E memory-chip facility in Taoyuan, Taiwan, Bloomberg reported Thursday, citing people familiar with the matter. The project, budgeted at $3 billion to $4.5 billion and scheduled to begin production in 2028, was to make DRAM and high-bandwidth memory (HBM) for AI data centers. Micron has informed Taiwanese officials and a supplier of the suspension, and will complete only limited safety work before freezing construction. The trigger is an escalating labor dispute: the Taoyuan union wants 15% of Micron's global operating profits distributed to employees (in cash), and Micron fears a deal reached with its larger Taichung union in October could be overturned. Micron said its existing Taiwan operations continue normally. Caveat: the pause is Bloomberg-sourced (people familiar with the matter) and unconfirmed by Micron — the Taoyuan union told China Times on Nov 5 the report is not official company information and needs confirming, and Taoyuan's city government said it had not received notice of a pause. Micron says its existing Taiwan operations continue normally.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nvidia has approached private equity firm Apollo Global Management about converting more than 130 telecom buildings owned by Apollo-backed broadband provider Brightspeed into a network of small AI data centers spanning as many as 20 U.S. states, The Information reported on Thursday, Nov 5. Nvidia submitted a written proposal to Brightspeed on Wednesday and has requested additional data and site visits. The sites together hold roughly 1 gigawatt of available power, which Nvidia estimates could support tens of billions of dollars' worth of its upcoming Vera Rubin servers, with retrofits potentially beginning in Q1 2027. Nvidia describes the proposed sites as an 'inference network' for running already-trained models rather than large-scale training clusters, with potential customers including SpaceX, Anthropic and OpenAI.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nvidia (NVDA): a workaround that keeps its chip deployments growing even if new mega-campuses are blocked
Apollo Global Management (APO): a potential new revenue stream for a distressed portfolio asset and a deepening tie to Nvidia's hardware cycle; hyperscalers and neoclouds (MSFT, AMZN, GOOGL, META, CoreWeave, Nebius, IREN): the distributed-inference model could ease their own siting constraints, though Microsoft separately declined to extend a major data-center agreement with Nvidia-backed startup Nscale; telecom carriers (AT&T, Verizon, Lumen): AT&T alone holds roughly 5,000 central offices, suggesting the retrofit playbook could extend far beyond Brightspeed's 130 buildings.
Apple raised about $7.5 billion from a seven-part U.S. dollar high-grade bond offering that priced Nov 5-6, its first bond sale in roughly 18 months (last: $4.5 billion in May 2025). The deal drew roughly $31 billion in orders, more than four times the final size. The offering comprises two floating-rate notes due 2028 and 2029 and five fixed-rate notes maturing 2028-2036, managed by Bank of America, Deutsche Bank, Goldman Sachs and JPMorgan. Pricing on the longest tranche (2036) tightened to a spread of just 0.55 percentage points over Treasuries, the tightest among recent Big Tech deals. Apple said proceeds are for general corporate purposes, including share buybacks, dividends and debt repayment — not AI infrastructure.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Apple (AAPL): cheap funding for buybacks/dividends without adding AI-capex leverage; the tight 0.55 ppt spread signals strong creditor confidence even in a selloff. Meta (META), Microsoft (MSFT), Alphabet (GOOGL), Oracle: their recent AI-linked deals priced wider, highlighting how the market is now differentiating between AI-capex debt and shareholder-return debt. Corporate bond market: a $31B order book in a risk-off session shows high-grade demand remains deep despite equity volatility.
South Korea's Kospi plunged by almost 9% at the market open on Friday, triggering a 20-minute trading halt on the exchange. The rout is the overnight follow-through from Thursday's Wall Street selloff, when the Nasdaq slumped 5.52% into correction territory, the S&P 500 fell 4.36% and the Dow dropped 1,833 points (3.44%) — wiping out roughly $2.5 trillion in market value in a single session. The Philadelphia Semiconductor Index fell 12.3%, its worst day in roughly six-and-a-half years, and the VIX surged about 70%, its biggest one-day jump in almost two years.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Cloudflare (NET) reported third-quarter 2026 results after Thursday's close that beat expectations across the board, sending shares up more than 20% in Friday premarket trading. Revenue rose 39% year over year to $782.5 million — the fastest quarterly growth in nearly four years and about 6% above the ~$737 million consensus — pushing annualized revenue past $3 billion. Non-GAAP EPS was $0.38 versus $0.34 expected, non-GAAP operating margin hit a record 17.5%, and current remaining performance obligations grew 37%. Management raised guidance for the third time this year: Q4 revenue of $845-846 million (vs ~$796 million consensus), full-year revenue of $2.963-2.964 billion, full-year non-GAAP EPS of $1.35 (vs $1.27 expected), and — most notably — GAAP profitability now expected as early as Q4 2026, pulled forward from a prior end-of-2028 target.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Cloudflare (NET) — surged more than 20% in Friday premarket trading, bucking Thursday's AI selloff; the beat-and-raise is a direct read on enterprise AI/security spending
AI infrastructure and cybersecurity peers — a demand signal that enterprise spending on AI-era network security is accelerating despite the broader AI-valuation rout; contrasts with the chip/hardware selloff
U.S. stocks closed sharply lower on Thursday, November 5, completing a violent two-day reversal after Wednesday's post-midterm rally. The Nasdaq Composite plunged 5.52% to 24,315.00, confirming correction territory (at least 10% below its October 6 record closing high) in its biggest single-day percentage drop since April 2025's 'Liberation Day' selloff. The Dow Jones Industrial Average fell 1,833 points, or 3.44%, to 51,523.06, fully erasing Wednesday's 1,500-point post-election surge. The S&P 500 dropped 4.36% to 7,353.73, with every sector finishing in the red. In dollar terms it was the largest single-day market-cap wipeout ever: S&P 500 firms shed $2.9 trillion and Nasdaq companies lost $2.5 trillion. The Magnificent 7 alone lost $1.04 trillion, led by Nvidia (-$374 billion, -9.3%) and Tesla (-$161 billion, about -10.8%), with Microsoft down $117 billion.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nvidia (NVDA): down ~9.3%, lost $374B in market value — the AI-chip leader is the epicenter of the data-center-slowdown fear. Tesla (TSLA): down ~10.8%, lost $161B — political/regulatory scrutiny of Musk's companies plus autonomy/AI plans under Democratic control. Microsoft (MSFT), Meta (META), Amazon (AMZN), Alphabet (GOOGL), Apple (AAPL): all fell, losing $117B, $58B, $55B, $42B and $59B respectively
Meta most exposed via data-center buildout in Louisiana, Ohio and Texas. Broadcom (AVGO) -10%, Qualcomm (QCOM) -9% on soft guidance, Arm (ARM) lower on in-line guidance. Neoclouds crushed: CoreWeave -10%, Nebius -15%, IREN and Applied Digital down more than 14%. Dell (DELL) extended its selloff, down ~7.5% Thursday. Chip ETF (SMH) posted its worst day since March 2020. Defensive sectors (health care, staples, energy, utilities) absorbed rotation flows.
October Employment Situation released Friday Nov 6 at 8:30 AM ET (BLS schedule confirmed, on time, no delay). Consensus: nonfarm payrolls ~+60-65k (Investopedia ~59k; Seeking Alpha 62k; several desks 65k; wide range 35k-125k) vs September's +29k; unemployment expected steady at 4.2% (range 4.1-4.3%); average hourly earnings +0.3% m/m (from +0.1%) and ~2.8-3.0% y/y (September 3.0%); labor-force participation expected flat at 61.8%. September's July/August revisions alongside. How to read it: a hot print (payrolls well above ~65k, unemployment ticking down, wages +0.4% or more) strengthens the December-hike case; a soft print does little to kill it while unemployment holds at 4.2% — the bar for the Fed to stand pat is real deterioration, not just another sub-100k month.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
The Nasdaq Composite fell more than 5% intraday on Thursday, November 5, pushing the index more than 10% below its recent record and into correction territory. The S&P 500 dropped about 3.3-3.4% and the Dow fell roughly 1,200 points (about 2.3%), wiping out most of Wednesday's post-midterm rally in a single session. The Philadelphia Semiconductor Index fell about 7.1%, and the iShares Semiconductor ETF dropped as much as 11% intraday, potentially its worst session since March 2020. Individual chip and AI-infrastructure names plunged: AMD fell about 13%, Qualcomm nearly 12%, Sandisk, Astera Labs, Seagate, Lumentum, Micron and Western Digital each fell 12-18%, Nvidia about 7% and Tesla about 9.5%.
Why it matters
The bigger picture behind this move, what it means for the company and the sector, and how it fits the trend we have been tracking for weeks.
What’s next
The dates, numbers and signals to watch from here, and what would change the story.
Why it matters & what’s next are part of Pro, along with the audio summary.
Nasdaq Composite and Nasdaq 100 (correction territory, down >10% from record highs); semiconductor sector broadly — Nvidia (NVDA, ~-7%), AMD (~-13%), Broadcom, Micron, Intel, Marvell, Qualcomm (~-12%), Sandisk, Seagate, Western Digital, Astera Labs, Lumentum
AI infrastructure/data-center names — Dell (DELL, extending its selloff), Arm, CoreWeave; hyperscalers — Meta (~-4%), Microsoft (~-2%), Amazon and Alphabet lower
Tesla (TSLA, ~-9.5%) on political and regulatory scrutiny of Musk-linked companies; defensive sectors (health care, staples, energy, utilities) absorbing rotation flows.
