South Korea’s record exports in September 2026 reached $120.9 billion, up 83.5 percent from a year earlier, driven by an AI-fuelled surge in chip shipments, Al Jazeera reports. The previous monthly record was $102.25 billion, set in June.
Chip shipments hit $60.3 billion in September, an increase of more than 260 percent, and made up nearly half of all exports. Al Jazeera notes that SK Hynix and Samsung Electronics control about 80 percent of the global high-bandwidth memory market and 60 percent of DRAM.
The year so far
Exports for January to September totalled $814.5 billion, against $709.7 billion for all of 2025. The trade surplus for the period was $49.85 billion. Nominal GDP grew 26.4 percent year on year in April to June, and the forecast for 2026 real growth is 3 percent, the strongest since 2017 outside the pandemic years, per the same report. The Kospi has risen nearly 60 percent this year.
Risks the government names
Trade Minister Kim Jeong-kwan called the result “a valuable achievement” but said global protectionism and Middle East tensions remain major variables. For the demand side, see our report on the Nvidia market value surge.
Nvidia stock jumped to a fresh intraday high on Friday, October 2, 2026, and the Nvidia market value surge pushed the chipmaker’s total worth above $5.7 trillion. The rally came the same day the U.S. government released a weak September jobs report. Soft hiring data usually rattles markets. This time, investors kept buying AI-related stocks instead, treating chipmakers as a safer bet for growth than the broader economy. Nvidia shares touched an intraday peak of $237.88, according to Yahoo Finance. The move extended a run that started a year earlier, when Nvidia first closed above $5 trillion in market value in October 2025. Friday’s figure marks a new, higher milestone, not a repeat of that earlier one.
Nvidia Market Value Surge By The Numbers
The broader market also had a strong day. The Nasdaq Composite rose 319.27 points, or 1.19%, to close at 27,190.86. The S&P 500 added 56.27 points, up 0.73%, to finish at 7,722.72. The Dow Jones Industrial Average gained 250.40 points, or 0.49%, to close at 51,176.96, according to Yahoo Finance. An intraday high reflects the peak price reached while markets are open, even if a stock closes lower by day’s end. Market capitalization, in turn, measures a company’s total value by multiplying its share price by its total number of outstanding shares — the yardstick investors use to judge company size.
Nvidia’s move stood out even against those gains. Hitting $237.88 a share pushed its total market capitalization past $5.7 trillion, a level no public company had reached before. Investors kept adding to AI-related positions throughout the session, and Nvidia led much of that buying. Because Nvidia ranks among the largest companies in the world by market value, its swings carry outsized weight across broad indexes like the Nasdaq and the S&P 500.
AI Chip Demand Fuels Nvidia’s Market Surge
Demand for AI chips has stayed strong through 2026, and that demand sits behind most of Nvidia’s gains this year. Data center operators, cloud providers and AI developers continue to order Nvidia’s chips faster than rivals can match capacity. That demand has also lifted companies that supply Nvidia’s ecosystem, from memory makers to software partners. Micron’s latest earnings pointed to a broader AI memory boom, a sign the chip rally extends well beyond Nvidia alone. Analysts and investors alike are watching how long that order backlog can hold up, since it underpins much of the optimism behind Nvidia’s valuation this year.
Nvidia has also worked to broaden its footprint beyond raw chip sales. Last week, the company launched an open agent safety platform aimed at AI developers, part of a broader push to stay central to how companies build and deploy AI systems, not just the hardware that runs them. That kind of move matters because it ties Nvidia’s business more tightly to software and services, not only to chip orders that can slow in any given quarter.
Friday’s $5.7 trillion figure builds on a milestone Nvidia reached almost exactly a year earlier. Nvidia first closed above $5 trillion in market value on October 29, 2025, according to a separate Yahoo Finance report from that date. That milestone followed comments from President Trump about discussing export approval for Nvidia’s Blackwell AI chips with CEO Jensen Huang.
Nvidia also announced a slate of deals at a Washington D.C. event around that time. The agreements included seven new U.S. Department of Energy supercomputers built with 10,000 Blackwell GPUs, a self-driving-car partnership with Uber, and agreements with Eli Lilly, Nokia, Oracle, Palantir and telecom firms on 6G research. Those deals touched government computing, autonomous vehicles, pharmaceutical research, telecom infrastructure and cloud software all at once, which showed how far Nvidia’s chips had spread beyond data centers. That breadth, visible a year before this week’s $5.7 trillion milestone, helped set the stage for the sustained demand that has carried Nvidia’s valuation higher since.
China Export Rules Still Shadow Nvidia’s Market Value
Chip export policy toward China remains an open question for Nvidia. In July 2025, the White House struck a deal that would allow Nvidia’s H20 chip to reach Chinese customers in exchange for a 15% revenue-sharing arrangement with the U.S. government. A revenue-sharing arrangement like this one would have Nvidia share a cut of certain China sales with the U.S. government in exchange for export approval — a structure not typically used in chip export policy. As of the October 2025 report, that arrangement had not been formalized, and Nvidia reported zero H20 sales to China in its most recent quarterly report at the time.
A year on, the underlying tension has not gone away. Export rules shape how much of the Chinese market Nvidia can actually reach, and Chinese chipmakers have not stood still while the rules stay unsettled. DeepSeek and Huawei have been building their own chip tools, an effort to reduce China’s reliance on Nvidia hardware regardless of how Washington’s export policy eventually settles.
Where Nvidia’s Rally Goes From Here
Friday’s rally came against a backdrop of economic uncertainty. The September jobs report showed nonfarm payrolls rose by just 29,000, far short of the 84,000 economists had expected, while unemployment held at 4.2%. That weak data pushed Treasury yields down, and investors moved into AI-related stocks that still look like a dependable growth story.
That pattern raises a real question: can AI-related stocks keep absorbing investor money every time other economic signals look shaky? For now, Nvidia’s position looks firm. Strong AI chip demand, a widening set of partnerships, and no sign of slowing orders all support the stock’s current levels. The unresolved China export question remains the clearest risk on the horizon, along with how long investors will keep treating AI stocks as a safe harbor from a cooling jobs market.
Nothing about Friday’s numbers guarantees where the stock goes next. Markets can reverse quickly, and a single trading session rarely settles a longer debate about valuation. What Friday did show is that, for now, AI chip demand has enough momentum to override a weak jobs report, at least for one of the world’s most closely watched stocks. Readers tracking the AI chip supply chain can follow the related threads above on Micron’s memory earnings, Nvidia’s new safety platform, and the Chinese chipmakers racing to catch up.
Nvidia’s Rally: Quick Answers
What pushed Nvidia’s market value past $5.7 trillion?
Nvidia shares hit an intraday peak of $237.88 on October 2, 2026, pushing its total market value above $5.7 trillion. Strong AI chip demand drove the move, even as a weak U.S. jobs report cooled other parts of the market.
When did Nvidia first reach a $5 trillion valuation?
Nvidia first closed above $5 trillion in market value on October 29, 2025, a year before this week’s $5.7 trillion milestone.
What drove Nvidia’s earlier $5 trillion milestone?
Comments from President Trump about discussing export approval for Nvidia’s Blackwell AI chips with CEO Jensen Huang helped drive that milestone, along with a slate of deals Nvidia announced in Washington D.C., including Department of Energy supercomputer orders, a self-driving-car partnership with Uber, and agreements with Eli Lilly, Nokia, Oracle, Palantir and telecom firms.
What is the status of Nvidia’s H20 chip sales to China?
In July 2025, the White House struck a deal allowing Nvidia’s H20 chip to reach China in exchange for a 15% revenue-sharing arrangement with the U.S. government. As of the October 2025 report, that arrangement had not been formalized, and Nvidia had reported zero H20 sales to China in its most recent quarterly report.
How did the broader stock market perform on October 2, 2026?
The Nasdaq Composite rose 1.19% to 27,190.86, the S&P 500 gained 0.73% to 7,722.72, and the Dow Jones Industrial Average rose 0.49% to 51,176.96.
Why did stocks rally despite weak jobs data?
The September jobs report showed nonfarm payrolls up just 29,000, versus 84,000 expected, with unemployment at 4.2%. That weak data pushed Treasury yields down, and investors shifted toward AI-related stocks as a safe haven for growth exposure.
CXMT G5 DRAM production is now underway in China. It marks one of the clearest signs yet that the country’s chipmakers are closing the gap with Samsung and SK Hynix. That progress comes despite years of export restrictions on advanced lithography gear.
Chinese memory chipmaker CXMT announced the milestone at the 2026 World Manufacturing Convention in Hefei on September 20. It unveiled its fifth-generation DRAM platform, known as the G5 Platform.
What CXMT G5 DRAM Production Actually Achieves
The G5 Platform uses quadruple patterning. That reaches an 11.95-nanometer active-area half-pitch for memory arrays. CXMT said its capacitor aspect ratio reaches 45:1. The core cell array height has shrunk to 6,762 nanometers. In plain terms, the company packs more memory into a smaller footprint. It says the platform yields at least 50% more dies per wafer than its predecessor.
How CXMT Got Here Without ASML’s Banned Machines
The achievement stands out for one reason. CXMT built it without access to ASML’s most advanced extreme ultraviolet lithography tools. Those remain subject to export controls. The company instead relied on quadruple patterning with existing equipment. That workaround took years to refine. It appears to be paying off at scale now.
Multiple patterning is not new in chipmaking. But matching the density gains EUV lithography delivers in a single pass usually costs more. It means more process steps, more time and more cost per wafer. CXMT chose to absorb that complexity rather than wait for policy changes. That choice reflects the broader bet Chinese chipmakers have made since export controls tightened.
What This Means for Samsung, SK Hynix and Global Memory Prices
CXMT now claims the title of China’s largest DRAM supplier. It also claims fourth place worldwide by market share. The company has scaled fast. It grew from roughly 40,000 wafers a month in 2020 to about 300,000 today. That output spans its 12-inch fabs in Hefei and Beijing. It targets 350,000 to 375,000 wafers monthly by year-end 2026. That capacity growth adds real supply to a global memory market squeezed by AI-driven demand.
The Products Already Shipping on the New Platform
Two 24-gigabit LPDDR5X products are already in mass production. They come in 496-ball and 245-ball packages for smartphones and portable electronics. Both are integrated into mainstream Chinese flagship handsets already, the company says.
The Bigger Picture for China’s Chip Ambitions
CXMT’s progress fits a wider Chinese push to build a self-sufficient semiconductor supply chain. That priority hardened after successive rounds of US export controls on advanced chipmaking equipment. Beijing has funneled heavy state-backed investment into domestic memory and logic chip production for years. CXMT’s rise from a minor player to a top-four global DRAM supplier is one of the clearest results so far.
Whether the company can keep closing the gap at the cutting edge remains an open question. Both Korean rivals keep investing heavily in their own next-generation nodes. Neither shows signs of slowing down.
Which Chinese Phone Makers Are Buying In
CXMT has not publicly named every customer using the new LPDDR5X chips. Industry watchers tracking teardown reports say several major Chinese smartphone brands have already qualified the parts for mainstream, non-flagship models. That matters commercially. Winning approval in China’s huge domestic phone market gives CXMT steady volume, even before it competes for premium international designs.
Analysts say the bigger prize is qualification with global device makers outside China. That would require CXMT to match not just density but also long-term reliability data that Samsung and SK Hynix have built up over decades. CXMT has not said when, or whether, it plans to pursue that qualification process.
Where CXMT Goes From Here
Capacity is still climbing toward its year-end target. CXMT’s next test is sustaining yields at scale while narrowing the technology gap further. Industry watchers expect the company’s progress to keep pressuring global memory pricing. That dynamic is already squeezing device makers worldwide this year.
The next milestone to watch is server and data-center memory. Can CXMT push the G5 platform into that higher-capacity segment? It remains dominated even more heavily by Samsung, SK Hynix and Micron than the mobile market CXMT has targeted so far. Analysts see that as the real test of whether CXMT can compete at the very top of the market, not just at the budget end.
Quick Answers on China’s New Memory Chip
What is CXMT G5 DRAM production?
It is the mass production of CXMT’s fifth-generation DRAM platform, announced at the 2026 World Manufacturing Convention in Hefei on September 20.
How advanced is the new chip?
The G5 Platform achieves an 11.95-nanometer active-area half-pitch using quadruple patterning, without relying on ASML’s most advanced banned lithography machines.
What products use it already?
Two 24-gigabit LPDDR5X chips, packaged for smartphones and portable electronics, are already in mass production and integrated into mainstream Chinese flagship phones.
How much is CXMT producing?
The company has grown from roughly 40,000 wafers per month in 2020 to about 300,000 wafers per month across its Hefei and Beijing fabs, targeting 350,000 to 375,000 by year-end 2026.
Where does CXMT rank globally?
CXMT now claims to be China’s largest DRAM supplier and the world’s fourth-largest by market share.
Broadcom shares fell after the company issued fourth-quarter guidance that trailed Wall Street’s target. The Broadcom revenue forecast miss overshadowed a blowout third quarter built on soaring AI chip sales. Broadcom told investors to expect roughly $34.8 billion in fiscal fourth-quarter revenue. Analysts polled by LSEG had expected about $35.03 billion, according to Reuters. The gap is small in percentage terms. But it landed on a stock priced for perfection, and traders sold first and asked questions later.
The Broadcom revenue forecast miss, by the numbers
Broadcom’s own numbers tell two different stories. The third quarter, reported on September 2, was excellent. Revenue hit $29.6 billion, up 86% from a year earlier, according to the company’s investor relations release. Non-GAAP operating income reached $20.1 billion, a 92% jump. Free cash flow came in at $13.7 billion, or 46% of revenue.
The trouble sits in the outlook. Broadcom guided fourth-quarter revenue to about $34.8 billion. That still marks 93% growth from a year earlier. Yet it fell short of the $35.03 billion consensus tracked by LSEG. Bloomberg’s own survey put the average estimate near $35.1 billion, with some analysts modeling above $36 billion. Broadcom also guided non-GAAP operating margin to about 66% of revenue for the quarter, flat against the year-ago period. Wall Street had penciled in further margin expansion, not a plateau.
What Hock Tan and Amie Thuener Told Investors
Broadcom CEO Hock Tan pointed to a specific growth engine. “Demand for our custom AI accelerators and networking continues to be very strong,” Tan said in the earnings release. He noted that AI semiconductor revenue reached $16.7 billion in the third quarter, up 221% year-over-year and 54% from the prior quarter. Tan told investors that momentum would continue. He guided AI semiconductor revenue to climb to $21.7 billion in the fourth quarter, a 236% annual increase.
CFO Amie Thuener echoed that confidence while acknowledging the flat margin outlook. “We delivered non-GAAP operating income growth of 92% year-over-year, as consolidated revenue grew 86% year-over-year to $29.6 billion,” Thuener said. She added that fourth-quarter revenue growth is forecast at 93% year-over-year, with the non-GAAP operating margin held at 66%, flat from a year ago. Investors read that as a signal. Heavy investment, not weakening demand, is compressing near-term profitability.
AI Chip Demand Is Soaring — So Why the Forecast Gap?
Broadcom’s custom AI chip business, often called XPUs, has become central to its growth story. The company designs application-specific processors for hyperscale customers, including Alphabet’s Google. Google’s tensor processing units, built with Broadcom, sit at the center of that partnership. Broadcom has also helped arrange financing so Anthropic can buy the computing power its custom chips help produce. That build-out sits inside a wider debate over chip export bills touching Anthropic and the industry group ITI. Export strength elsewhere reinforces the point. South Korea’s chip exports hit a record in the same stretch, a sign that global AI infrastructure spending has not slowed.
Competition is the other half of the story. Rival Nvidia delivered a strong forecast of its own the week before Broadcom reported. Bloomberg Intelligence analysts said Nvidia had set a high sales bar for the rest of the sector. Marvell Technology and MediaTek are both racing to win custom chip work, and both count Nvidia as a backer. Marvell announced a deal last month to help build custom semiconductors for Google, a customer Broadcom has long served. That competitive pressure, more than any drop-off in AI demand, appears to explain the cautious guidance.
How Wall Street Reacted to the Forecast Miss
Broadcom shares dropped about 4% in after-hours trading once the numbers landed, according to Bloomberg. The stock had climbed roughly 6% for the year heading into the report, leaving little room for disappointment. Some analysts pushed back on the sell-off. Bernstein’s Stacy Rasgon said the underlying quarter was strong. He called the results very, very good and noted that semiconductor revenue actually landed slightly above expectations. Other analysts framed the move as a reset in expectations after a run of AI-driven gains, rather than a sign of fading demand.
What Comes Next for Broadcom’s AI Chip Business
Broadcom’s own guidance points to continued expansion, not contraction. Management has said it expects AI-related revenue to keep growing sharply into fiscal 2027 and 2028, based on commitments already on the books. Broadcom is not alone in wiring multi-year AI compute commitments into today’s numbers. OpenAI has been in its own funding talks tied to a rising valuation, and its backers are locking in the same kind of long-term computing bets Broadcom is chasing with Anthropic. The financing arrangements with Apollo Global Management and Blackstone still need to close and fund actual chip purchases, a process that will take years to play out. Investors will also watch whether new custom chip customers emerge beyond Google, and whether Broadcom can defend its lead against Marvell and MediaTek. The next real test comes when Broadcom reports fourth-quarter results in December. The market will then see whether $34.8 billion in revenue landed, or beat it.
Frequently Asked Questions
What is the Broadcom revenue forecast miss?
The Broadcom revenue forecast miss refers to the company’s fourth-quarter guidance of about $34.8 billion, below the $35.03 billion analysts expected, according to LSEG data cited by Reuters.
Why did Broadcom stock fall after strong earnings?
Shares fell because the fourth-quarter outlook missed consensus estimates, even though third-quarter revenue grew 86% and beat expectations. Investors focused on the forward guidance rather than the results already reported.
How much did Broadcom’s AI chip revenue grow?
Broadcom’s AI semiconductor revenue reached $16.7 billion in the third quarter, up 221% from a year earlier. The company guided AI semiconductor revenue to $21.7 billion for the fourth quarter.
What operating margin did Broadcom guide for the fourth quarter?
Broadcom guided non-GAAP operating margin to approximately 66% of revenue for the fourth quarter, flat compared with the year-ago period.
Is competition affecting Broadcom’s custom chip business?
Yes. Marvell Technology and MediaTek are both pursuing custom AI chip deals, and Nvidia’s own strong forecast the week before Broadcom’s report raised expectations across the sector.
Does the forecast miss mean AI chip demand is slowing?
Not according to Broadcom’s own numbers. Guidance still calls for 93% year-over-year revenue growth in the fourth quarter. Analysts including Bernstein’s Stacy Rasgon described the underlying business as strong.
Positron AI has closed an $875 million Series C at a $5 billion post-money valuation, and the Positron AI funding round is a bet on a specific contrarian idea: that the memory bottleneck in AI inference can be solved with cheap commodity chips instead of the scarce, expensive high-bandwidth memory everyone else is fighting over.
The company announced the raise in two tranches: $375 million in the Series C proper, and a Series C-1 of up to $500 million.
The bet: skip HBM entirely
Inference workloads are memory-hungry. Serving a large model means holding enormous weights close enough to the compute to be read fast, which is why high-bandwidth memory has become the industry’s tightest constraint — supply is limited, advanced packaging capacity is limited, and Nvidia absorbs much of both.
Positron’s next-generation chip, Asimov, pairs its compute architecture with 288GB to 2,304GB of LPDDR5X per chip. LPDDR5X is commodity memory, the kind that goes into phones and laptops. It is slower per bit than HBM and vastly more available.
The wager is that for inference specifically, capacity beats peak bandwidth — that fitting more of the model in memory matters more than reading any single part of it at maximum speed. If that holds, Positron sidesteps the supply chain that constrains its competitors.
It is an engineering claim, not yet a proven one, and the company is the party asserting it.
Who wrote the cheques
The round was co-led by NEA, Atreides Management, Valor Equity Partners, Andra Capital, Dylan Patel’s SemiAnalysis Capital, and Jim Clark — the founder of Silicon Graphics and Netscape.
Additional investors include the Qatar Investment Authority, DFJ Growth, Cisco Investments, Hudson River Trading and Naver Ventures, alongside existing backers.
Two names stand out for what they signal. SemiAnalysis Capital is the investment arm attached to a research operation that scrutinises exactly this kind of architectural claim, which makes its participation a form of technical endorsement. Hudson River Trading is a high-frequency trading firm — a buyer whose entire business depends on inference latency, not a generalist fund.
Where the money goes
Per the company, the financing funds three things: the Asimov tapeout; a 2MW-plus engineering data centre and emulation platform; and the production ramp of Titan, its current-generation inference system, including LPDDR5X supply commitments and go-to-market expansion.
Asimov is scheduled to tape out on TSMC’s N3P process at the end of 2026, with production targeted for the second half of 2027, as SiliconANGLE reported.
That timeline is the thing to hold onto. Silicon announced in 2026 for production in late 2027 has roughly eighteen months of execution risk ahead of it, in a market where the competitive baseline moves every quarter.
Titan is what bridges the gap. It is the current-generation system, shipping now, and the LPDDR5X supply commitments funded by this round are as much about securing memory allocation for that ramp as about the future chip. Commodity memory is abundant relative to HBM, but “abundant” is not the same as “available at volume on contract”, and locking supply early is how a company avoids becoming the constraint it was designed to escape.
The competitive picture it enters
Inference silicon has become the most contested corner of the AI hardware market, because it is where the recurring revenue is — training happens in bursts, inference runs continuously.
The incumbents are not standing still. Qualcomm committed to a large-scale inference partnership with Amazon, covered in our report on that deal, and Nvidia’s manoeuvring in the space has drawn antitrust attention, as we set out in the DOJ’s look at the Groq arrangement.
A $5 billion valuation for a company whose flagship product ships in eighteen months prices in a lot of confidence. The structure of the raise — a second tranche of “up to” $500 million — suggests the investors built themselves some optionality about how much of that confidence to fund immediately.
The question that decides it
Whether commodity memory is genuinely good enough for production inference at scale, or good enough only for workloads where latency tolerance is generous.
Positron says the former. Its customers will establish which is true, and the evidence arrives in 2027. Until then, the round is a well-capitalised hypothesis.
Questions about the raise
How much did Positron raise and at what valuation?
$875 million total at a $5 billion post-money valuation, split into a $375 million Series C and a Series C-1 of up to $500 million.
What is different about the Asimov chip?
It uses commodity LPDDR5X memory — 288GB to 2,304GB per chip — rather than high-bandwidth memory, avoiding the HBM and advanced packaging supply constraints that limit competing inference hardware.
When will Asimov be available?
It is scheduled to tape out on TSMC’s N3P process at the end of 2026, with production targeted for the second half of 2027.
Who led the round?
Co-leads were NEA, Atreides Management, Valor Equity Partners, Andra Capital, SemiAnalysis Capital and Jim Clark. Other investors include the Qatar Investment Authority, Cisco Investments and Hudson River Trading.
Is Positron competing directly with Nvidia?
In inference, yes. It is not targeting training workloads, where Nvidia’s position is strongest.
What is the main risk?
Execution and timing. The flagship chip is roughly eighteen months from production in a market where competing hardware advances continuously, and the performance claim for commodity memory is not yet demonstrated at scale.
