Broadcom Inc

Bank of America estimates Meta Platforms could save roughly $8.5 billion in 2027 by running AI workloads on its own custom silicon instead of buying third-party chips, an outside analyst estimate rather than company guidance. The figure rests on a specific roadmap: Meta plans to deploy its third-generation MTIA 450 chip, code-named Arke, in the first half of 2027, followed by the higher-performance MTIA 500, or Astrid, later that year, both co-developed with Broadcom and aimed at AI inference workloads. BofA models Meta deploying 5 to 6 gigawatts of owned capacity in 2027 at a total cost of roughly $200 billion, assumes chips make up 60% of that spend, and pegs Meta's custom silicon as about 40% cheaper than third-party equivalents. Broadcom CEO Hock Tan said custom chips optimized for a customer's own workloads outperform any GPU and can do so at half the cost, and confirmed Broadcom will deliver three generations of MTIA accelerators to Meta between now and the end of 2027. Meta's FY2026 capex guidance sits at $130 billion to $145 billion, narrowed from $125 billion to $145 billion, with total expense guidance raised to $165 billion to $169 billion, while Q2 2026 revenue reached $60.80 billion, up 27.96% year over year, on advertising revenue of $59.36 billion.

Broadcom CEO Hock Tan has guided the company's annual AI semiconductor revenue to reach $230 billion in its fiscal 2028, up from $16.7 billion in the third quarter, an annualized pace of just over $65 billion. Tan's earlier forecast, made in late 2024, projected AI revenue of $60 billion to $90 billion in 2027, but Broadcom now projects $115 billion in AI revenue for that year, making the original guidance a major underestimate. Broadcom's custom AI chip business, known as application-specific integrated circuits, counts Alphabet, Meta Platforms, OpenAI, and Anthropic among its clients. Tan said the company has secured the supply chain necessary to meet that demand. Wall Street analysts estimate Broadcom's total revenue will be $106 billion this fiscal year, leaving $41 billion in non-AI revenue if the AI projection holds.

Broadcom CEO Hock Tan said on Mad Money on Sep. 14 that Anthropic CEO Dario Amodei's call for an intentional slowdown in advanced AI development does not change Broadcom's AI semiconductor forecasts. Amodei published an essay titled "We Must Pace the Frontier" on Sep. 12, 2026, drawing endorsements from OpenAI CEO Sam Altman and Elon Musk, and AVGO fell over 4.8% on Sep. 14 while the iShares Semiconductor ETF dropped 5.6%. Tan said demand for compute infrastructure for AI development, frontier models and inference remains very strong and durable, agreed that AI needs governance and safeguards, but pushed back on the framing that AI poses existential risk at current capability levels, saying it is not a live animal that will run wild by itself. Broadcom reported Q3 fiscal 2026 AI semiconductor revenue of $16.7 billion, up 221% year-over-year and 54% quarter-over-quarter, with total consolidated revenue of $29.6 billion, up 86% year-over-year, and free cash flow of $13.7 billion, or 46% of revenue. Q4 guidance calls for AI semiconductor revenue of $21.7 billion, up 236% year-over-year, and consolidated revenue of $34.8 billion, up 93% year-over-year, while full fiscal 2026 AI revenue is now expected to be $58 billion, raised from prior guidance of $56 billion. Tan guided fiscal 2027 AI semiconductor revenue to approximately $115 billion, doubling to approximately $230 billion in fiscal 2028, and reiterated confidence in exceeding $30 in EPS by fiscal 2028, with Anthropic on track to become Broadcom's largest XPU customer in 2027-2028 with visibility to 10 gigawatts of deployment by 2028. Tan also helped establish a $35 billion AI special-purpose vehicle platform with Apollo and Blackstone to support Anthropic's one-gigawatt deployment.

Broadcom is trading at just 19 times forward earnings even as its AI semiconductor revenue surged 221% year over year to $16.70 billion, a disconnect that 24/7 Wall St. calls an opportunity with a $423.95 price target implying 22.31% upside from the current $344.05 quote. The call follows a fiscal Q3 report in which total revenue reached $29.591 billion, up 85.5% year over year, with non-GAAP EPS of $3.32 beating consensus, and management guided Q4 AI revenue to $21.7 billion. CEO Hock Tan said Q3 demand was simply hot and that the company is just getting started, and management now targets fiscal 2027 AI revenue of $115 billion and fiscal 2028 AI revenue of $230 billion across a $350 billion two-year shipment pipeline spanning six XPU customers including Google, OpenAI, Meta, and Anthropic. Broadcom's 19x forward multiple looks cheap against NVIDIA at 24x, AMD at 33x, and Marvell at 56x, despite faster AI revenue acceleration than those peers, and the Wall Street consensus target sits at $531.85 with 37 Buy and 8 Strong Buy ratings. Execution risk remains the biggest concern, with Tan flagging land, power, and shell constraints on capacity deployment and potential bottlenecks in HBM memory, substrates, and leading-edge wafers, while the bear scenario lands at $374.92, still above today's price.

Anthropic agreed to spend $517 billion on computing power over the 11 months through the end of August, according to The Information, far above the $180 billion in commitments it had previously disclosed. The spending is set to benefit Nvidia, whose GPUs remain the primary chips powering AI workloads and which could invest up to $10 billion in Anthropic as an anchor investor in its IPO. Advanced Micro Devices will invest up to $5 billion in Anthropic, which will buy AMD's Helios rack-scale system and deploy up to 2 gigawatts of AMD GPUs beginning in 2027. Anthropic is also set to become Broadcom's largest customer as it deploys Tensor Processing Units that Alphabet co-designed with Broadcom, with Broadcom saying Anthropic will deploy 1 gigawatt of Alphabet's Ironwood TPUs this year, 5 gigawatts of its TPU v8i in 2027 and another 10 gigawatts in 2028. Amazon, which owns around a 20% stake in Anthropic, has received more than $100 billion in AWS cloud commitments from the lab over the next decade.

Broadcom, Micron Technology and Marvell Technology all raised forward guidance in their latest earnings reports, citing accelerating AI infrastructure demand. Broadcom reported fiscal Q3 2026 revenue of $29.59 billion, up 85.5% year over year, with AI semiconductor revenue of $16.7 billion, up 221% year on year, and guided Q4 AI semiconductor revenue to $21.7 billion, up 236% year over year, outlining a trajectory reaching approximately $115 billion in fiscal 2027 and $230 billion in fiscal 2028. Micron posted fiscal Q3 2026 revenue of $41.46 billion, up 345.7% year over year, and guided fiscal Q4 revenue to $50.0 billion plus or minus $1.0 billion at gross margin near 86%, with full-year fiscal 2026 capital spending of approximately $27 billion. Marvell reported Q2 FY2027 revenue of $2.739 billion, up 36.5% year over year, and raised its fiscal 2027 data-center growth outlook to approximately 60% from approximately 50% previously, with CEO Matt Murphy flagging the October 6, 2026 Investor Day as the next catalyst. Micron's fiscal Q4 report, expected late this month, is the next test of whether hyperscaler capital spending holds through year-end.

Meta Platforms is moving deeper into custom AI silicon, with a new generation of internally designed chips set to enter its data centers in the first half of 2027. The company is currently testing its third-generation MTIA 450 processor, code-named Arke, while its successor, MTIA 500, or Astrid, is expected to complete design work in about a month and reach data centers by the end of 2027. Meta is working with Broadcom on chip design and Taiwan Semiconductor Manufacturing Co. on production, and has committed to deploying more than a gigawatt of the chips over a 12-month period. Twelve Arke chips delivered by TSMC on Sept. 1 performed within 2% to 3% of Meta's simulations and have already run Meta models alongside models from DeepSeek and Alibaba. Meta also canceled its planned Olympus processor, which was intended to handle both AI training and inference, partly because of cost concerns, and is instead prioritizing inference, the day-to-day running of AI models.

Four major infrastructure vendors have now shipped standalone agent governance products at general availability, a rush that has itself become the signal that agent governance is decoupling from individual platforms to become a category of its own. Okta pushed furthest into new territory with its July 2026 product innovations, shipping Agent-to-Agent Connections at general availability to enable secure multi-agent workflows through temporary runtime tokens that enforce which agents may invoke which others, alongside the Agent Gateway, available as a research release, which sits between agents and the systems they access without requiring code changes. IBM's Think 2026 announcement positioned next-generation watsonx Orchestrate as an agentic control plane, introduced in June on AWS and IBM Cloud, offering runtime policy management, credential health monitoring, and an Agent Access overview across an organization's entire agent estate. Broadcom embedded governance directly into the compute fabric with AgentMinder, unveiled at VMware Explore on August 31 and shipping at general availability bundled into the VMware Private AI Cloud, treating agents as enterprise-grade identities bound to a declared mission, permitted intents, approved tools, and authorized resources. Dataiku made a different architectural bet, with Dataiku Agent Management scanning agents across nine platforms including Microsoft Copilot Studio, Salesforce Agentforce, AWS Bedrock, and Google Vertex to provide a cross-platform control tower for discovery, certification, and audit-readiness. The urgency tracks to two numbers: Menlo Ventures found that 76 percent of AI applications are purchased rather than built internally, and the Cloud Security Alliance reported in February that 84 percent of organizations doubt they could pass a compliance audit focused on agent behavior or access controls.

Broadcom's AI semiconductor sales surged 221% year over year to $16.7 billion in its third quarter, pushing total revenue to $29.59 billion and lifting AI to about 56.4% of quarterly sales. Management guided fourth-quarter revenue to roughly $34.8 billion, including about $21.7 billion from AI chips, which would raise AI to roughly 62.4% of revenue and imply another $5 billion of sequential AI-chip growth in a single quarter. The stock slipped about 0.6% to $342.50 Tuesday morning as chip stocks struggled to rebound from Monday's AI-driven selloff, leaving shares 14.89% below GuruFocus' GF Value estimate of $402.41. The muted reaction reflects investor focus on how long hyperscalers can sustain their current spending pace.

Nvidia is in talks to become an anchor investor in Anthropic's initial public offering, a listing that could raise as much as $100 billion at a valuation of around $2 trillion, with Nvidia reportedly considering an investment of up to $10 billion. Anthropic's annualized revenue run rate has reportedly climbed from about $9 billion in 2025 to more than $65 billion by mid-2026, and the company is targeting as much as $200 billion of revenue by 2028. The strategic stakes go beyond financial return: Anthropic has committed $30 billion to Microsoft Azure infrastructure powered by Nvidia chips and is reportedly committing $45 billion to rent AI computing capacity from Nscale, infrastructure expected to use Nvidia's Vera Rubin chips. Nvidia has previously disclosed an investment and technology partnership with Anthropic, and its fiscal 2026 filing said it had entered into an agreement to invest up to $10 billion in the company, while Nvidia generated $215.9 billion of fiscal 2026 revenue, up 65%, with Data Center revenue up 68%. The risk is that Anthropic is diversifying its hardware base, with partnerships spanning Amazon, Google and Broadcom, custom chip development, and a recent agreement to buy billions of dollars of AI servers from AMD, meaning Nvidia's investment could help finance a customer that is reducing its dependence on Nvidia hardware. Anthropic's valuation had already reached $965 billion following its May 2026 financing, so the prospective IPO would represent a more than doubling of valuation in a relatively short period, and Nvidia's fiscal 2026 gross margin was already 71.1%, down from the prior year, partly because of its transition toward full-scale Blackwell data-center systems and a $4.5 billion charge related to H20 inventory and purchase obligations.

Piper Sandler initiated coverage of Broadcom Inc. with an Overweight rating and a $460 price target on September 9, with analyst David O'Connor calling the company the leader in ASIC chips with a 75% share of the ASIC market for AI inference. O'Connor estimates demand is running twice the available supply, making component availability and deployment timing the key determinants of whether that demand converts into earnings. Broadcom is already ramping Meta's MTIA program and OpenAI's Jalapeno chips along with two other customers, while Google's TPU program remains its major volume ASIC program and Anthropic scales deployments on AVGO-enabled TPU compute; a networking attach rate of around 30% could generate an estimated $20-30 billion of total content per gigawatt of deployed capacity. Piper sees a line of sight to 12 gigawatts of demand in fiscal 2027, rising toward an estimated 38 gigawatts by fiscal 2030, underpinning roughly 51% EPS compound growth through the decade, and its $460 target is based on a 14x multiple of estimated fiscal 2028 earnings. The firm also flags risks including eroding pricing power, customer concentration and lower relative gross margins, while 170 hedge funds held the stock in the second quarter, down slightly from 173 in the prior one.

Broadcom has purchased the Irvine office compound it previously rented, paying former landlord PRP Real Assets $325 million for the 660,000-square-foot research campus near the junction of the 5 and 405 freeways. The deal ranks among the largest office sales in Southern California in recent years and marks a reacquisition for Broadcom, which developed the property as an office and research-and-development complex before selling the campus for roughly $443 million in 2017 and agreeing to lease it for 20 years. PRP acquired the campus for $355 million in 2020. The campus consists of two five-story office buildings at 15101 and 15191 Alton Parkway, completed in 2017 and 2018 and designed to be Broadcom's corporate headquarters with significant research-and-development capabilities. Broadcom, now based in Palo Alto, is one of the world's largest semiconductor and infrastructure software companies, and through acquisitions has assembled a portfolio of 17 owned properties totaling more than 2.2 million square feet, according to CoStar data. Companies occupying their own buildings accounted for 34% of Orange County office sales volume over the past year, compared with a five-year average of 24%, CoStar said.

President Donald Trump used a Truth Social post on Monday morning to call Anthropic CEO Dario Amodei "Dario (Anthropic!), who is now pretending to be a 'perfect little angel,'" while asserting that "the only control or 'guardrails' that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT." In the same message, Trump said "We already have tremendous CRIMINAL and REGULATORY power over these companies!" and framed opposition to the buildout as "a SICK conspiracy going on against AI and Data Centers, and the only one that is happy about it is China." The escalation follows Amodei's warning that advanced AI could wipe out half of entry-level white-collar jobs and his repeated calls for a slower deployment cadence. The clearest listed beneficiaries of a light-touch posture are picks-and-shovels names, with NVIDIA reporting Q2 FY27 revenue of $96.22B, up 105.8% year over year, and data center revenue of $89.02B, up 117%, and guiding Q3 to $108.0B plus or minus 2%. Broadcom, the custom-accelerator play, posted Q3 FY26 AI semiconductor revenue of $16.7 billion, up 221% year on year, and guided AI revenue to $21.7 billion in Q4, with CEO Hock Tan saying Anthropic is deploying one gigawatt of Ironwood in 2026, another five gigawatts of TPU v8i in 2027, and an incremental 10 gigawatts of line-of-sight capacity in 2028. Anthropic has no US-listed common equity, and the ticker ANTW is a themed ETF tracking an AI basket rather than a direct way to invest in Anthropic on this news.

Broadcom and Nvidia have both reported earnings for their most recent quarters, with Nvidia's adjusted revenue growing 106% year over year and diluted earnings per share rocketing 120%, while Broadcom posted adjusted net revenue growth of 86% and adjusted diluted EPS growth of 96%, and both companies topped Wall Street average estimates. On guidance, Broadcom projected revenue of $34.8 billion in its current quarter, slightly below analyst estimates of $35.03 billion, but Chief Executive Officer Hock Tan said on the earnings call that the company believes its AI semiconductor revenue is set to double in 2027 to $115 billion and then double again in 2028 to $230 billion, with the 2028 guidance coming in $50 billion ahead of analyst estimates. Nvidia forecast revenue of $108 billion in the current quarter, roughly $2.5 billion above estimates, and projected 70% annual revenue growth in fiscal year 2028, with most of that occurring in calendar year 2027, while analysts had only estimated 44%. On gross margin, Nvidia reported roughly 75% in its second quarter of fiscal 2027 and expects a 74% gross margin in the current quarter, narrowing to the 71% to 72% range as the fiscal year progresses, which management attributed to higher memory costs, while Broadcom had a consolidated gross margin of 75% and anticipates a slip to 73% in the current quarter as it sells more memory-intensive custom chips. Nvidia generated $89 billion in AI chip revenue last quarter, compared with Broadcom's nearly $21 billion, and Nvidia still trades at a cheaper forward earnings multiple than Broadcom.

Goldman Sachs chief U.S. equity strategist Ben Snider said on CNBC's Squawk on the Street that AI investment spending is driving about half of S&P 500 earnings growth. Snider noted the index traded at 22x earnings at the start of the year and now sits at 19x with the index near a high, and that the ten-year Treasury yield has risen to 4.83 percent from close to 4 percent a year ago. The growth leans on a small group of buyers: NVIDIA posted Q2 FY27 revenue of $96.22 billion, up 105.8 percent year over year, with data center at $89.02 billion, while Broadcom's AI semiconductor revenue reached $16.70 billion in Q3, up 221 percent year over year, with a Q4 outlook of $21.7 billion. Microsoft, on the buyer side, carries full-year capex of $115.95 billion and an FY27 capex expectation of roughly $175 billion, and its 365 Copilot reached over 30 million paid seats with commercial remaining performance obligations up 84 percent to $678 billion. Snider said Goldman's positioning indicator is at its lowest level since March, even with the market near a high, while September seasonality and the next CPI release add near-term risk with the VIX at 16.46 and the target rate at 3.75 percent.

Nvidia is partnering with Apollo Global Management, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to raise over $500 billion in third-party capital for AI infrastructure, as Bank of America warns the sector may need about $1.2 trillion of external finance to support AI capital expenditures forecast to exceed $5 trillion between 2026 and 2030. Bank of America says major chip suppliers are taking on an unexpected role as credit intermediaries, helping remove financial risks that make massive AI data centers difficult or expensive to finance through minimum revenue commitments, take-or-pay contracts, and residual value guarantees. Broadcom's AI XPV Platform has secured senior notes and the remaining value of chips covering $31 billion of an initial $35 billion loan package arranged with Apollo and Blackstone, nearly 87% of the debt package, with the guaranteed portion priced at 5.75% against 8.5% for an unsecured second lien. In August, Nvidia said it would provide finance assistance for land, electricity, and construction at SB Energy's PORTS-Pike Technology Campus in Ohio, with a first rollout of 4.25 gigawatts of AI factory capacity expected to be used by OpenAI, and Nvidia also said it was investing $1.5 billion in SB Energy. Broadcom on Sept. 2 reported $16.7 billion in AI semiconductor sales for its fiscal third quarter, up 221% from a year earlier, and CEO Hock Tan said the company expects AI semiconductor revenue of about $21.7 billion in its fiscal fourth quarter, an increase of 236% year-over-year.

Broadcom and AMD both reported blockbuster quarters that put them squarely in the conversation as challengers to Nvidia's dominance in AI compute. Broadcom's AI semiconductor revenue hit $16.7 billion, up 221% year over year and 54% sequentially, and the company guided for Q4 AI revenue of roughly $21.7 billion, up 236%, with fiscal 2027 AI revenue projected at approximately $115 billion. AMD's Data Center segment reached $6.72 billion, up 107% year over year, powered by MI350 shipments and record EPYC sales, with Anthropic committing to up to 2 GW of MI450 Series and Microsoft expanding Azure deployments. Broadcom CEO Hock Tan argued that co-designed custom silicon runs at less than half the cost of a GPU, while AMD's Lisa Su called it an excellent quarter with record revenue and profitability. Broadcom trades at a forward P/E of 20 and AMD at a forward P/E of 32, and both CEOs flagged HBM, substrates, packaging, and data center power as the key choke points for 2027.

Taiwan Semiconductor Manufacturing is trading at $428.29 against a 24/7 Wall St. price target of $492.37, implying 14.96% upside on a buy rating with 90% confidence. The call rests on record August 2026 revenue that surged 53.3% year over year, with January-through-August revenue up 39.3%, prompting management to raise full-year 2026 revenue guidance to slightly above 40% USD growth. Q2 2026 delivered EPS of $4.31 against a $3.8866 estimate, revenue of $40.20 billion, up 36.05% year over year, and a 67.7% gross margin, while Q3 revenue is guided to $44.6 billion to $45.8 billion and the 2026 capital budget was raised to $60 billion to $64 billion. The bull case leans on demand from Nvidia, whose Q3 guide of $108 billion, and Broadcom, whose Q3 FY26 AI semiconductor revenue reached $16.7 billion, up 221% year over year, effectively lock in TSMC's advanced-node order book, while Intel Foundry lost $2.1 billion in its latest quarter. The main near-term risk is gross-margin compression, with the 2nm ramp expected to dilute gross margin by 3 to 4 percentage points and overseas fabs adding another 2 to 3 points of drag, alongside Taiwan-Strait geopolitics as an unpriced tail risk; the bear-case one-year price is $413.79 against a bull case of $525.08 and a sell-side consensus target of $552.38.

OpenAI is deepening its partnership with Samsung Electronics to co-develop next-generation chips, a move aimed at advancing its enterprise AI services. Harrison Kim, General Manager of OpenAI Korea, said at a press conference in Seoul on Wednesday that the joint production and research on the next-generation chips is one of the areas where OpenAI has gained the most recognition and made the most progress with Samsung Electronics, according to Reuters. Kim also said Samsung is one of the largest-scale deployments of ChatGPT globally, with the AI used across departments including research and development, marketing and sales, and he highlighted growing demand for memory chips to support faster, more advanced computing. In October 2025, OpenAI enlisted Samsung Electronics and SK Hynix to supply chips and equipment for its Stargate project, while Samsung SDS, Samsung C&T and Samsung Heavy Industries explored potential partnerships with OpenAI on floating data centers and advanced infrastructure design. OpenAI has also been expanding its custom chip push, launching its first custom Intelligence Processor, Jalapeño, with Broadcom in June, and CFO Sarah Friar said at Goldman Sachs' Communacopia + Technology Conference in San Francisco on Monday that the company used its own AI models to develop the Jalapeño chip, which was finalized for production within nine months.

Broadcom's AI semiconductor revenue surged to $16.7 billion in its fiscal third quarter, up 221% year over year, with management guiding to $58 billion for fiscal 2026 and $230 billion by fiscal 2028. The company's Jalapeno chip, OpenAI's first-generation custom accelerator built by Broadcom, runs frontier AI workloads at half the cost of a GPU, a cost advantage that management argues makes it indispensable to hyperscalers as borrowing costs climb. Broadcom reported total Q3 revenue of $29.59 billion, up 85.5% year over year, with free cash flow of $13.67 billion, equal to 46% of revenue, and guided Q4 AI revenue to $21.7 billion, up 236% year over year. Chief Executive Hock Tan said on the call the company is on target to exceed $30 in earnings per share in fiscal 2028, nearly five times the fiscal 2025 figure of $6.82, while the dividend was raised 10% in December 2025, the fifteenth consecutive annual increase since fiscal 2011. The company carries $23.98 billion in cash against $59.6 billion of gross fixed-rate debt at a 4% weighted coupon, and repurchased $7.8 billion of stock in the first quarter of fiscal 2026 under a $10 billion authorization approved in March 2026, though customer concentration among Google, Anthropic, OpenAI, and Meta remains a risk.

Mark Zuckerberg's net worth surged by $13 billion in a week as Meta Platforms stock extended its rally, climbing 8% from August 31 to September 4 and lifting his fortune from $206 billion to $219 billion. The gains follow Meta's release of Muse Spark 1.3, an AI model that scored comparably to top rivals from OpenAI and Anthropic, and the launch of Meta Compute, a cloud business selling excess compute to outside customers. Meta also plans to double its infrastructure capacity to 14 gigawatts next year and is starting production on its own custom AI chip, Iris, built with Broadcom. Despite the rally, Meta shares remain down about 7.1% year-to-date through September 8, and Zuckerberg's net worth is still below its level at the start of the year. The company faces ongoing risks, including legal issues and heavy capital expenditures, but revenue grew 28% year over year to $60.8 billion in Q2 2026, and the stock trades at just 20 times forward earnings.

OpenAI CFO Sarah Friar confirmed at Goldman Sachs' Communacopia conference on September 8 that the company used its own frontier AI models to design the Jalapeno custom inference chip, marking the first confirmed instance of an AI lab using its own intelligence to architect its silicon. The chip, co-developed with Broadcom and announced in June 2026, achieved a nine-month design-to-tape-out cycle, the fastest ever reported for high-performance hardware, though the figure is company-claimed. This move is tied to OpenAI's aggressive pricing, including an 80% reduction for the Luna model to $0.20 per million input tokens and $1.20 per million output tokens, which has driven a 10x usage increase and supports 32% enterprise revenue growth from June to July 2026. By controlling the silicon stack, OpenAI aims to decouple margins from third-party cloud providers, reinforcing its position as a full-stack utility ahead of its confidential S-1 filing with an $852B post-money valuation. Broadcom CEO Hock Tan indicated small prototypes are expected in late 2026, with full-scale ramp slated for the first half of 2028, a period that will test the scalability of this recursive design model.

Artificial intelligence (AI) technology has begun to generate real applications and revenue at the enterprise level, creating a chain of positive impacts from software and cloud to semiconductors. Yuanta Securities noted that AI adoption is transforming software costs into cloud revenue, with companies like Snowflake, Salesforce, and ServiceNow benefiting from AI agents, while NVIDIA is expanding its moat by controlling the source through Hugging Face. However, SEMICON Taiwan 2026 highlighted bottlenecks, such as TSMC's demand for chip-making equipment nearly doubling in six months, and memory issues that forced Google to reuse DDR4. The Asian supply chain is fully benefiting, with South Korea's semiconductor exports surging 209% year-over-year. Earnings from Dell, Broadcom, Snowflake, Credo, and Palo Alto Networks reflect strong demand, such as Dell's record-high AI server orders at $60.9 billion and Broadcom's AI revenue growing 221%.

Shares of Applied Materials, AMD, Broadcom, Lam Research, and Marvell Technology surged in afternoon trading after Qualcomm announced a multi-generational collaboration with Amazon Web Services to develop custom AI data center chips, and ASML, TSMC, and Intel revealed key advancements in next-generation chip manufacturing. Qualcomm's partnership with AWS focuses on co-designing custom AI chips for high-performance inference and data center workloads, aiming to cut operating costs and improve energy efficiency. At the SPIE Photomask conference, ASML and TSMC launched an initiative to transition from 6-inch to 12-inch photomasks, targeting a pilot line by 2031 and full production by 2033, which would reduce costs and boost output for advanced AI processors. Additionally, Intel Foundry and ASML confirmed that Intel has processed over one million wafers on its Intel 18A node using High NA EUV equipment. These developments boosted investor confidence in sub-2-nanometer chip designs and AI infrastructure, with Applied Materials up 3.4%, AMD up 6.5%, Broadcom up 3.3%, Lam Research up 4.2%, and Marvell Technology up 3.3%.