Meta’s Muse AI agent drives CPU demand surge, stock up 25% in two weeks
Meta launched its consumer AI agent Muse on September 8, quickly topping the US App Store free chart with 2.8 million downloads in 12 days. The product assigns each user a dedicated cloud virtual machine, shifting compute demand toward CPUs. Meta’s stock surged over 25%, adding more than $400 billion in market cap. CPU-focused chipmakers Arm, Intel, and AMD rallied, with AMD briefly reaching a $1 trillion market cap. Analysts forecast the CPU-to-GPU ratio in data centers could shift from 1:8 to 1:1, benefiting the entire computing supply chain including IC substrates.
Editorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page reads the event directly, while its address stays stable when the title changes.
- Summary covers the current reports
Cross-source coverage
Common ground
- All agree that Meta's Muse is not a proven business model yet, with concerns about user retention and revenue projections.
- There is agreement that the AI compute shift from GPU-heavy to CPU-heavy processing is a plausible trend, supported by supply chain data like Intel's 50% order fulfillment rate.
- All recognize that the debate involves significant geopolitical and economic power dynamics, not just technical or market factors.
Points of contention
- The Neutral Agent argues that supply chain data like Intel's fulfillment rate is the key driver, while the Eastern and Regional Agents insist that politics and power structures shape those supply chains.
- The Eastern Agent defends China's AI ecosystem as sovereign and lawful, while the Neutral and Regional Agents criticize it as a tool for state control and digital colonialism.
- The Regional Agent calls for community-owned and open-source alternatives, but the Neutral Agent says such solutions are unrealistic without massive capital and infrastructure.
Blind spots
- All overlook the environmental and labor costs of the AI boom, such as energy consumption and factory working conditions, as raised by the Regional Agent.
- The debate focuses on Western and Chinese power, but ignores the voices and needs of communities in the Global South who are not part of the decision-making.
- No one addresses whether the hyper-connected, surveillance-heavy model of AI is what people actually want, rather than just assuming it's inevitable progress.
WorldAttention’s read
This roundtable shows that while there's some agreement on the technical shift toward CPU-centric AI and the hype around Muse, the real divide is over who controls the technology and for whose benefit. The Neutral Agent sees supply chain constraints as the main story, but the Eastern and Regional Agents argue that politics, sovereignty, and human impact matter more. A key blind spot is that the Global South is treated as a market or data source, not a creator of its own digital future. Ultimately, the debate reveals that without addressing power imbalances, labor conditions, and genuine local agency, the conversation is just about which empire or corporation gets to dominate.
Reporting timeline
China Gold Securities: Meta's Muse AI Agent Accelerates New Computing Narrative, Benefiting Entire Supply Chain
China Gold Securities (国金证券) released a research report stating that Meta's launch of the independent AI agent application Muse has driven a new narrative for computing power. Muse quickly topped the App Store charts in the US and Canada, boosting Meta's stock price by 21% in two weeks. The report argues that Muse marks the transition of the agent concept to a tangible product, expanding the potential user base from millions of developers to billions of consumers, increasing the demand pool by 20-100 times. This shift ushers in the 'inference era,' with token consumption expected to surge. Muse reconfigures hardware ratios, shifting the CPU-to-GPU ratio from 1:4-1:8 to 1:1-1:2, leading to a structural revival for CPUs. The report forecasts that the entire computing supply chain—including GPUs, memory, SSDs, and network infrastructure—will benefit from the expanded token scale. Additionally, Meta is integrating Muse into AI glasses and a portable voice hardware device called Muse Charm, creating diversified hardware entry points for the agent. Risks include user growth and commercialization falling short of expectations, security and privacy incidents, CPU supply and price volatility, platform ecosystem resistance, and trade friction.
Read sourceSinolink Securities: Meta's Muse AI Agent Accelerates New Compute Narrative, Benefiting Supply Chain
Sinolink Securities released a research report stating that Meta's independent AI agent application, Muse, has rapidly gained downloads, topping the App Store charts in the US and Canada and boosting Meta's stock price by 21% in two weeks. This marks the transition of the agent narrative from concept to actual product. Targeting billions of consumers, Muse expands the potential demand pool by 20-100 times, ushering in the inference era and significantly increasing token consumption. The report argues that Muse restructures the ratio of compute hardware, shifting the CPU-to-GPU ratio from 1:4-1:8 to 1:1-1:2, driving a structural revival for CPUs. This shift benefits the entire compute supply chain, including GPUs, memory, SSDs, and network infrastructure. Sinolink also notes that Muse empowers edge hardware like AI glasses and the Muse Charm device, creating diverse hardware entry points for the agent. Risks include slower-than-expected user growth, security incidents, CPU supply and price volatility, platform ecosystem resistance, and trade friction.
Read sourceMeta's Muse AI Agent Sparks CPU Demand, Analysts Flag IC Substrate as Next Growth Sector
This article from Zhitong Finance reports that Meta's new AI agent, Muse, has ignited a new narrative in the AI market, shifting focus from model development to product and scenario-driven demand. At Meta Connect, Muse was central to all new AI hardware releases. Meta CEO Mark Zuckerberg stated that Muse is core to a future where AI belongs to everyone, predicting it could become a super-intelligence with billions of users. Unlike previous AI agents, Muse operates on a dedicated cloud virtual machine for each user, allowing tasks to continue in the background. This has driven up CPU-related stocks like Arm and Intel, with AMD hitting a record high. Analysts from Jefferies and Puyin International Securities forecast that the surge in CPU demand will cascade to the IC substrate supply chain, particularly ABF (Ajinomoto Build-up Film) substrates, which have a near 1:1 relationship with server CPU shipments. They note a multiplicative effect: each CPU requires one FCBGA-type ABF substrate, and higher performance demands will increase substrate layers and size. South Korea's Meritz Securities reports that Samsung Electro-Mechanics has signed a long-term MLCC contract with a North American CPU maker and is increasing ABF substrate shipments. Taiwan's Kinsus Interconnect Technology reported August revenue of NT$4.81 billion (approx. RMB 1.02 billion), up 40.6% year-on-year, driven by high-end ABF substrates. Kinsus noted strong demand, with ABF capacity utilization at 85% and expected to reach 95% by year-end, and some clients are already negotiating capacity allocations for 2027-2029. The company plans to increase ABF monthly capacity by 25% by 2027, focusing on high-end AI substrates.
Read sourceShow 12 older updatesHide older updates
Meta's Muse AI Agent Drives CPU Demand, Analysts See IC Substrate as Next Growth Link
This article analyzes the market impact of Meta's new AI Agent, Muse, which was unveiled at the Meta Connect conference. Muse is distinguished by its ability to run tasks in the background on a dedicated cloud virtual machine for each user, lowering the barrier for mainstream AI adoption and driving a surge in CPU demand. This has boosted stocks of CPU suppliers like Arm, Intel, and AMD. Analysts from Jefferies and浦银国际证券 (SPDB International) forecast that this demand will cascade to the IC substrate market, particularly ABF (Ajinomoto Build-up Film) substrates used in high-end server CPUs. They note a near 1:1 relationship between server CPU shipments and ABF substrate demand, with additional demand from substrate specification upgrades. Evidence of this trend includes strong earnings from substrate maker景硕科技 (Kinsus), which reported 40.6% revenue growth driven by high-end ABF substrates, and plans to increase ABF capacity by 25% by 2027. The article also cites a report from Korea's Meritz Securities indicating increased ABF substrate shipments from Samsung Electro-Mechanics to a North American CPU maker.
Read sourceMeta Muse AI Agent Sparks CPU Demand, IC Substrate Seen as Next Growth Link
An article from East Money News, citing a report from the科创板日报 (Science and Technology Board Daily), analyzes the market impact of Meta's new AI Agent, Muse, launched at the Meta Connect conference. Muse is described as a super-intelligent AI that lowers the barrier for ordinary users by integrating browser operations, account connections, payments, and background execution. A key feature is that each user has an independent cloud virtual machine, allowing tasks to continue even when the app is closed, which has driven CPU demand and stock prices for companies like Arm and Intel. Analysts from Jefferies and浦银国际 (SPDB International) forecast that this demand will cascade to the IC substrate market, particularly ABF (Ajinomoto Build-up Film) substrates used in FCBGA packaging for server CPUs. They note a near 1:1 ratio between server CPU shipments and ABF substrate demand, with additional demand from substrate specification upgrades. Evidence of strong demand includes a 40.6% year-on-year revenue increase for景硕科技 (Kinsus Interconnect Technology) in August, driven by high-end ABF substrates, and plans to increase ABF production capacity by 25% by 2027. The report also mentions Samsung Electro-Mechanics supplying ABF substrates to a North American CPU maker.
Meta Muse AI Agent Sparks CPU Demand, Analysts Flag IC Substrate as Next Growth Sector
Meta's new AI agent, Muse, has ignited market enthusiasm, with analysts viewing it as a catalyst for a new wave of AI demand driven by products and scenarios. At Meta Connect, all new AI hardware was tied to Muse, which CEO Mark Zuckerberg said could become a super-intelligence serving billions. Muse's unique feature—a dedicated cloud virtual machine per user that continues tasks in the background—has boosted CPU demand, lifting stocks like Arm, Intel, and AMD. This demand is expected to ripple up the supply chain to IC substrates, which serve as the critical interface between chips and circuit boards. Jefferies noted Muse signals AI agents entering the mainstream consumer market, driving demand for CPUs, PCBs, and ABF substrates. Puyin Securities forecast a transmission chain from user expansion to task frequency to CPU and substrate demand, with a near 1:1 ratio between server CPU shipments and ABF substrate demand. Samsung Electro-Mechanics has secured a long-term MLCC contract with a North American CPU maker, and its ABF substrate shipments are rising. Kinsus Interconnect Technology reported August revenue up 40.6% year-on-year, driven by high-end ABF substrates, with utilization expected to rise to 95% by year-end. Some clients are already negotiating capacity allocation for 2027-2029, and Kinsus plans to increase ABF monthly capacity by 25% by 2027, focusing on high-end AI substrates.
Read sourceMeta's Muse AI Agent Drives CPU Demand, Analysts See IC Substrate as Next Growth Link
Meta's new AI Agent, Muse, has sparked market excitement, with analysts predicting it will drive a new wave of AI demand expansion. Unlike previous agents, Muse assigns each user a dedicated cloud virtual machine, enabling persistent background task execution, which has boosted CPU demand and lifted stocks like Arm and Intel. Multiple brokerages, including Jefferies and Pu Yin International Securities, forecast that this CPU demand will cascade to IC substrates, particularly ABF (Ajinomoto Build-up Film) substrates used in server CPUs. Pu Yin notes a near 1:1 mapping between server CPU shipments and ABF substrate demand, with additional upgrades in substrate layers and size. South Korea's Meritz Securities reports Samsung Electro-Mechanics increasing ABF substrate shipments to a North American CPU maker. Taiwan's Jing Shuo Technology reported August revenue up 40.6% year-on-year, driven by high-end ABF substrates, with capacity utilization expected to rise from 85% to 95% by year-end. The company plans to increase ABF monthly capacity by 25% by 2027, focusing on high-end AI substrates.
Meta's AI Agent Muse Sparks Global Frenzy, Boosts Stock and Investment Hopes
Meta's AI agent 'Muse' has gone viral globally within two weeks of launch, driving Meta's stock up over 25% and adding more than $400 billion to its market capitalization, which now approaches $2 trillion. Cathie Wood's ARK funds bought over $30 million in Meta shares on September 21, while multiple investment banks have raised Meta's price target. JPMorgan raised its target from $820 to $920, citing Muse's monetization shift from subscriptions to transaction fees. Unlike the earlier open-source AI agent OpenClaw, which faded due to security risks, Muse is a hosted service that runs tasks in a cloud virtual machine, lowering deployment barriers. Analysts from China International Capital Corporation note that security architecture is central to Meta's narrative. However, concerns remain about third-party integration, privacy, and user trust. Some analysts, including Oppenheimer's Jason Helfstein, question whether Muse will generate significant profit gains. The article notes that AI application benefits are currently flowing to hardware, with AMD's stock rising 10% on September 21. Bernstein argues that market fears about agents disrupting platforms are ahead of actual fundamental changes.
Read sourceMeta Muse AI Agent Sparks CPU Demand, Brokerages Flag IC Substrate as Next Growth Sector
Meta's Muse AI agent, unveiled at Meta Connect, is viewed by analysts as a catalyst for a new wave of AI demand driven by products and scenarios. Muse's unique feature of a dedicated cloud virtual machine per user, enabling persistent background task execution, has boosted CPU-related stocks like Arm, Intel, and AMD. Brokerages including Jefferies and SPDB International forecast that the surge in CPU demand will cascade to the IC substrate supply chain, particularly ABF substrates. Jefferies notes Muse signals AI agents entering the mainstream consumer market, driving demand for CPUs, PCBs, and ABF substrates. SPDB International outlines a transmission path from user expansion to task execution load to CPU and substrate demand, with a non-linear amplification effect due to the near 1:1 ratio of server CPUs to ABF substrates and ongoing specification upgrades. Meritz Securities reports Samsung Electro-Mechanics has signed an MLCC long-term contract with a North American CPU maker and is increasing ABF substrate shipments. Substrate maker Kinsus Interconnect Technology reported August revenue up 40.6% year-on-year, driven by high-end ABF substrates, with capacity utilization expected to rise from 85% to 95% by year-end. Some customers are already negotiating capacity allocation for 2027-2029, prompting Kinsus to plan a 25% increase in ABF monthly capacity by 2027, focused on high-end AI substrates.
Read sourceMeta's Muse AI Agent Sparks Investment Frenzy, Driving Stock Surge and Analyst Upgrades
Meta's newly launched personal AI agent, Muse, has gained rapid global adoption, topping the U.S. App Store free charts within two weeks and accumulating 2.8 million downloads. The product's success has driven Meta's stock up over 25% since its September 8 release, adding more than $400 billion in market value and pushing the company's valuation toward $2 trillion. Notable investor Cathie Wood's ARK funds added over $30 million in Meta shares, while multiple investment banks including JPMorgan, Deutsche Bank, Wells Fargo, and Jefferies raised their ratings or price targets, citing Muse's potential as a consumer AI 'killer app.' Unlike the open-source OpenClaw agent, Muse offers a managed, secure environment with independent virtual machines and user-confirmed sensitive operations, lowering deployment barriers. However, analysts caution that sustained success depends on task completion reliability, user retention, and cost coverage. The article also notes that AI application gains are currently benefiting hardware sectors like CPUs and PCBs, with some Wall Street analysts questioning Muse's profitability and potential disruption to platform businesses.
Read sourceMeta's Muse AI Agent Sparks Investment Frenzy, Stock Surges 25% in Two Weeks
Meta's newly launched AI agent, Muse, has gone viral globally within two weeks, driving Meta's stock price up over 25% and adding more than $400 billion to its market capitalization, which now approaches $2 trillion. On September 21, Meta shares surged 11.43%, their biggest single-day gain in over a year. Cathie Wood's ARK Investment Management bought over $30 million worth of Meta shares on September 21 through three ETFs. Multiple investment banks have raised Meta's rating or target price; Morgan Stanley raised its target from $820 to $920, citing Muse's monetization model shifting from subscriptions to transaction fees. The article compares Muse to the earlier OpenClaw 'lobster' agent, which faded due to security risks, noting Muse's hosted cloud virtual machine approach lowers deployment barriers. Analysts from China International Capital Corporation highlight security architecture as a core narrative. Some Wall Street analysts, including Oppenheimer's Jason Helfstein, question whether Muse will generate significant incremental profits, citing concerns about willingness to pay, competition, and user trust. The article notes that AI application benefits are flowing first to hardware, with AMD shares also rising. Bernstein argues market concerns about agents disrupting platforms are ahead of actual fundamental changes.
Read sourceMeta's Muse AI Agent Drives CPU Demand, AMD and Intel Stocks Surge
Meta's personal AI agent product Muse topped the US App Store free chart, with over 902,000 downloads in six days. Muse is designed as a persistent AI agent that runs continuously in a dedicated cloud virtual machine (2 vCPU, 8GB RAM, 100GB storage) even when the app is closed, only pausing for user approval on sensitive actions. Analysts from Barron's and Jefferies forecast that widespread adoption of such AI agents will significantly increase AI infrastructure workloads, particularly boosting demand for server CPUs. This reverses the recent trend of GPU dominance in AI data centers. Intel and AMD are seen as primary beneficiaries. On September 21, AMD shares rose 9.95% (market cap over $1 trillion) and Intel shares rose 12.14% (market cap $643.74 billion). Industry projections suggest the CPU-to-GPU ratio in data centers could shift from 1:8 or 1:4 to 1:1. Arm estimates CPU core demand per gigawatt of data center compute could quadruple to 120 million cores. At 100 million Muse users, demand could reach 2 billion vCPU cores, though real-world sharing and idle management could reduce this significantly.
Read sourceMeta Muse AI Agent Drives CPU Demand Surge, Analysts See Structural Shift in Compute
Meta's AI agent Muse, launched September 8, reached 1.8 million US iOS downloads in 13 days, outpacing ChatGPT's 1.3 million. Its daily active users in the US hit 642,000, triple ChatGPT's同期 figure. The product assigns each user a dedicated cloud virtual machine with 2 vCPUs, driving CPU demand. Meta's stock rose 11.43% on September 21, adding ~$190 billion in market cap; Arm, Intel, and AMD also surged. Analysts argue agentic AI workloads shift compute balance from GPU to CPU. Bernstein estimates CPU share rises from 14% to 50% in agent tasks. Arm predicts CPU core demand quadrupling to 120 million per GW. Intel's server CPU supply is at ~50% of orders, with a planned 10% PC CPU price hike. KeyBanc reports Intel and AMD's 2026 server CPU capacity nearly sold out. Chinese brokerages see 2026 as a breakout year for domestic CPUs like Haiguang and Kunpeng. However, Muse's monetization model (free + $20/$100 subscriptions) and Oppenheimer's 2027 revenue estimate of $27.5 billion depend on 191 million users and 6% conversion, which remain unverified.
Read sourceMeta's AI Agent Muse Sparks CPU Rally, Analysts See Structural Shift in AI Compute Demand
Meta's personal AI agent Muse, launched on September 8, surpassed 2.5 million downloads in six days and topped Apple's US App Store free chart, igniting a rally in chip stocks. AMD's market cap exceeded $1 trillion, while Intel and Arm surged. Analysts at Wedbush, Morgan Stanley, Goldman Sachs, and Jefferies argue that AI agents shift compute bottlenecks from GPU to CPU and memory, potentially adding $32.5-60 billion to the server CPU market by 2030. Morgan Stanley notes that as AI moves from generation to autonomous action, CPU orchestration becomes more strategic than raw GPU power. However, Oppenheimer's Jason Helfstein cautions that Meta would need ~115 million Muse subscribers at $20/month to generate $27.5-28 billion in annual revenue, which he deems unlikely due to conversion and trust issues. Amazon has blocked Muse from its retail site over data concerns. The article notes the rally is sentiment-driven on an unverified order chain.
Read sourceMeta's Muse AI Agent Drives Nasdaq Surge, CPU Stocks Rally, Micron Earnings in Focus
Meta's consumer AI agent Muse, launched on September 8, reached 448,000 daily active users and 264,000 single-day downloads within 12 days, far exceeding ChatGPT's early adoption. This triggered a 'quiet surge' in the Nasdaq 100, which rose 2.83% on September 21 to reclaim 30,000 points, its highest since June. However, Goldman Sachs analyst Peter Callahan noted the rally was unusually quiet with low trading activity, concentrated in mega-cap tech stocks. CPU-focused chipmakers Arm Holdings, Intel, and AMD surged over 10-15%, as analysts argue AI agents require more CPU processing than GPU-heavy generative AI. AMD briefly reached a $1 trillion market cap. The rally was driven by options activity rather than fundamentals, and credit markets showed divergence, with investment-grade CDS spreads widening to five-month highs. Investors expressed caution, citing macro headwinds from oil prices and interest rates. Next week's Micron Technology earnings are seen as a key test for the sustainability of the AI infrastructure trade, especially as storage stocks lagged behind CPU gains.
Read source