Meta Platforms launched its new personal AI agent, Muse, which became the most-downloaded free iPhone app in the U.S. over the past three days.
Meta Platforms Inc. and Vital Knowledge
The global semiconductor market experienced a sharp upswing as investors reassessed the underlying compute architecture required for consumer-facing artificial intelligence. The catalyst was Meta Platforms Inc.’s Muse AI agent, which climbed to become the most-downloaded free application in the U.S. Meta launched Muse two weeks ago as a virtual personal assistant that can book appointments and trips, place orders, and send emails and messages on behalf of users. Meta () surged 11% during Monday’s trading session, helping pull the Nasdaq-100 up 2.15% as part of a broader market advance. Unlike traditional conversational chatbots designed primarily for answering questions, Muse operates as an agentic assistant capable of performing multi-step tasks in the background, such as sending emails, booking travel, placing orders, and managing schedules on behalf of users. Market intelligence firm Vital Knowledge emphasized to clients that while initial generative AI tasks relied primarily on GPU acceleration for text and image generation, autonomous agents require continuous background routines, logic loops, and local data coordination. App Adoption
Sensor Tower and Wells Fargo
The early adoption trajectory for Muse outpaced several historical benchmarks for consumer software. Data from Sensor Tower cited by Wells Fargo analyst Ken Gawrelski documented significant user growth following the app’s independent release. Peak Downloads: The app recorded 264,000 single-day downloads in the U.S. on September 19, marking its highest single-day total and sustaining a streak of downloads over multiple days for three consecutive days. Daily Active Users: Muse reached 448,000 daily active users by its tenth day on September 18. Historical Contrast: OpenAI’s ChatGPT took nearly a year to surpass daily U.S. downloads and 49 days to reach daily active users, according to the sourced figures. Gawrelski noted that Muse Assistant and Instinct signal the transition from chatbots to consumer AI assistants, emphasizing that early market positioning is critical because assistant utility scales with usage time. Bolstered by these metrics, Wells Fargo raised its price target on Meta shares from $640 to $796 while maintaining an Overweight rating and adjusting 2027 earnings per share multiple assumptions to reflect positive signals from a new major product cycle. Wall Street and Global Chip Stocks Rally on CPU Demand
Arm Holdings Plc and Samsung Electronics Co.
The enthusiasm surrounding agentic AI triggered broad gains across hardware sectors that supply central processors and memory components. Market analysts at Vital Knowledge told clients that while initial generative AI workloads relied heavily on graphics processing units for text and image creation, autonomous agents require architectures that heavily favor central processing units. In U.S. trading, Arm Holdings Plc shares surged about 16% to lead the Nasdaq-100 pack, while Intel Corporation shares rose 13% and Advanced Micro Devices Inc. climbed 9%. Advanced Micro Devices Inc. also topped $1 trillion in market value. The Philadelphia Stock Exchange Semiconductor Index closed up 4.3% on Monday, marking its largest single-day rally since August 4. The momentum carried over into Asian markets. South Korean semiconductor and equipment suppliers rose sharply as Samsung Electronics Co. climbed as much as 3.5% and SK Hynix Inc. gained up to 3.3%, lifting the benchmark Kospi Index by 2.3%. Both South Korean manufacturers supply high-bandwidth memory crucial for global AI hardware infrastructure. In morning trading, SK Hynix and Samsung Electronics shares rose as much as 3.6% and 3.5%, respectively, outperforming the benchmark Kospi’s 1.6% gain. Kyle Rodda, a senior analyst at Capital.com, noted that these developments may provide additional support for cyclical areas of the market. Arm Holdings and Meta Target Agentic Infrastructure
Arm Holdings and Meta
The hardware rally intersects directly with structural shifts in data center design. Meta serves as Arm’s lead partner and co-developer on custom central processing units built specifically for agentic artificial intelligence infrastructure. Arm introduced its data-center CPU in March, noting that application workloads driven by agents increase compute requirements by at least fourfold. Meta previously stated that its data centers are outgrowing the operational capacity of traditional processors as AI workloads expand, and noted in April that compute requirements are evolving to demand more CPU as agentic AI work advances. Arm Holdings stated that CPU requirements are increasing by at least fourfold as application workloads become more agent-driven. While Meta is Arm’s lead partner and co-developer on the AGI CPU built for agentic AI infrastructure, neither company has said that Meta’s Muse runs on Arm’s AGI CPU. Arm is working with Meta across multiple chip generations, with the chips intended to operate alongside Meta’s own MTIA accelerators. Investors and analysts are watching for further hardware and software announcements as Meta CEO Mark Zuckerberg prepares to deliver a keynote address at the annual Meta Connect conference on Wednesday night.
Keep reading