Meta Muse AI Agent Sparks Chip Stock Rally, AMD Hits $1T

On Monday, September 21, Wall Street handed Meta Platforms its sharpest single-day rally in more than a year — and advertising had almost nothing to do with it. A two-week-old personal AI agent called Muse, built to send emails, book appointments and shop on a person’s behalf, had climbed to the top of Apple’s free app charts in the U.S. Investors read that as proof Meta finally has a credible consumer AI product, and the excitement did not stop at Meta’s stock. It spilled into the chip sector, driving Advanced Micro Devices past a $1 trillion market valuation for the first time and pushing Intel and Arm into double-digit gains. By Tuesday morning, some of that enthusiasm had already started to cool — a reminder of how quickly sentiment can swing in the current AI trade.

Smartphone screen showing AI assistant app icons including ChatGPT, Meta AI and Gemini, representing the competitive AI agent app market — Photo by Salvador Rios on Unsplash

What Happened?

Meta Platforms closed Monday’s session at $741.25, up 11.43% on the day, with trading volume running well above its three-month average. The move followed a note from Wells Fargo analyst Ken Gawrelski, who raised his price target on Meta from $640 to $796 and pointed to early consumer demand for Muse, the personal AI agent Meta launched earlier this month, as evidence the company now has a genuine AI story to tell heading into its Meta Connect developer conference this week.

The rally quickly widened beyond Meta itself. Semiconductor stocks jumped as traders bet that a shift toward autonomous AI agents like Muse would require significantly more computing capacity, not just for training large models but for running them continuously on behalf of millions of users. AMD gained roughly 10% and closed above a $1 trillion market capitalization for the first time, becoming just the 14th U.S. company to reach that milestone. Intel rose around 12%, and Arm Holdings jumped more than 17%. The Philadelphia Semiconductor Index climbed about 4.3%, its fifth consecutive day of gains.

Key Details

What Muse Actually Does

Meta introduced Muse on September 8 as a personal AI agent designed to handle everyday digital chores rather than simply answer questions. According to Meta, the tool can send emails, book travel, fill out forms, cancel or lower recurring bills, and shop online, connecting to a user’s calendar, payment methods and other apps only after the person opts in. It runs on Muse Spark, a family of models Meta built specifically for multistep, agentic tasks, and Meta’s head of product for its Superintelligence Labs, Nat Friedman, has said the product drew heavy inspiration from OpenClaw, an open-source personal-agent project that circulated widely among developers earlier this year.

Within roughly two weeks of launch, Muse had briefly topped the overall free-app chart on Apple’s U.S. App Store and reached the number-two spot among Productivity apps, according to rankings cited by multiple trackers. Sensor Tower data reported by CNBC and other outlets showed U.S. iOS downloads climbing from the tens of thousands at launch into the hundreds of thousands within two weeks, with one industry tracker estimating the app had surpassed 2.5 million downloads in its first 13 days — ahead of comparable agent downloads from OpenAI’s ChatGPT and Anthropic’s Claude in that same window.

Friction With Amazon

Not every company has welcomed Muse. Amazon has blocked the agent from operating on its retail site after Meta declined a request from the e-commerce giant to remove the bot, according to Bloomberg. The move fits a pattern: Amazon has restricted access to outside AI shopping and browsing agents for more than a year, arguing that unsanctioned bots threaten the roughly $56 billion advertising business built around shoppers browsing directly on Amazon.com. Meta, meanwhile, has been striking its own retail partnerships, including deals that would let Muse book restaurant reservations through OpenTable and, eventually, check out using Shopify’s Shop Pay.

Why Chipmakers Rallied

The chip rally reflects a broader wager among investors: that agentic AI tools performing tasks continuously in the background — rather than answering a single prompt and stopping — will consume far more computing power, particularly during inference, the stage where a trained model is actually put to work generating responses and taking actions. While GPUs remain central to training and running large models, agentic systems also lean heavily on general-purpose CPUs for orchestration, data handling, networking and enterprise integration, which is where AMD, Intel and Arm compete most directly.

That demand narrative already had momentum. Intel’s chief executive, Lip-Bu Tan, said last week that the company could currently meet only about half of customer demand for its chips. Arm has estimated the server CPU market could grow more than 35% annually to reach $120 billion by 2030, while AMD has projected even faster growth, forecasting that server CPUs could expand by roughly 50% through 2030 into a $220 billion revenue opportunity. Muse’s early traction gave investors a concrete, consumer-facing example to point to.

Stock market candlestick chart on a dark trading screen, representing the chip stock rally following Meta Muse AI agent's early success — Photo by Maxim Hopman on Unsplash

Why This Matters

For most of 2026, investor enthusiasm around AI has centered on GPU makers and the hyperscalers spending tens of billions of dollars on data centers, with comparatively little evidence that ordinary consumers were adopting agentic AI tools at scale. Muse’s download numbers, however preliminary, gave the market its clearest signal yet that a mainstream audience will actually install and use an AI agent that takes real actions — sending money, booking travel, managing subscriptions — rather than just chatting. That distinction matters because it is the difference between AI as a curiosity and AI as infrastructure with a genuine revenue path, which is exactly the kind of proof point Wall Street has been waiting for.

It also reframes the competitive map. Meta has spent much of 2025 and 2026 fielding questions about whether its enormous AI capital spending — guided at $130 billion to $145 billion for 2026, most of it tied to AI infrastructure — would ever translate into a product people actually want. Muse’s early reception, even without disclosed revenue or retention figures, gives Meta something it has lacked: a visible, mainstream AI product that isn’t Facebook or Instagram’s news feed.

What It Could Mean for AI Users and Businesses

For everyday users, Muse adds to a fast-growing category of AI agents built to act rather than just answer, joining a field that already includes offerings from OpenAI and Anthropic. That raises real questions about privacy and account access that go beyond convenience, since granting an agent access to email, payments and calendars is a materially bigger trust decision than typing a question into a chatbot.

For merchants and platforms, Amazon’s decision to block Muse signals that the fight over how AI reshapes online shopping is intensifying rather than settling. Retailers now have to decide whether to welcome AI agents that could bring in sales but bypass their own advertising and search placements, or block them and risk missing out if agentic shopping becomes mainstream. Businesses evaluating which assistant to standardize on may also want to revisit how Muse stacks up against the options compared in Claude, ChatGPT and Gemini, since download charts alone say little about reliability, security or cost for business use.

For investors and the broader tech sector, the episode is a preview of how quickly capital can rotate toward “picks and shovels” plays — chipmakers, cloud infrastructure, networking — whenever a consumer AI product shows real traction, even before that traction converts into disclosed revenue.

What Meta and Analysts Are Saying

Meta has not disclosed daily or weekly active users for Muse, nor an incremental capital-expenditure figure tied specifically to the product, and the company’s broader 2026 capex guidance was not revised upward in connection with the launch. Wells Fargo’s Gawrelski based his price-target increase on download data and on the view that heavy usage could eventually push meaningful numbers of users toward Muse’s paid tiers. Other analysts have urged caution about durability, noting that download charts are volatile and that Muse’s privacy model — which routes the agent through an isolated cloud environment Meta calls a “Secure VM” — has not yet faced sustained real-world scrutiny.

What to Watch Next

Meta Connect, September 23: Meta’s annual developer conference is expected to bring further Muse updates and could offer the first hints of monetization plans.
Retention data: Whether Muse’s download surge converts into sustained daily use, rather than a short-lived novelty spike, will determine if the chip-demand thesis holds up.
Retailer responses: Whether other e-commerce platforms follow Amazon’s lead in restricting AI shopping agents, or instead pursue integrations similar to Meta’s deals with OpenTable and Shopify.
Chip supply: Whether Intel, AMD and Arm can actually meet the higher demand forecasts investors are now pricing in, given Intel’s own acknowledgment of existing supply constraints.

Conclusion

Monday’s rally was less about a single earnings beat than about a narrative investors have been waiting to see proven: that agentic AI can attract real, mainstream demand, and that demand has knock-on effects across the technology supply chain. Muse’s download numbers are still early, unaudited by independent researchers, and unaccompanied by usage or revenue data from Meta itself. Whether the moment marks a genuine turning point for agentic AI adoption or simply a sharp, sentiment-driven trading day will likely become clearer only after Meta Connect and a few more weeks of download and retention data. For now, it has already reshaped, at least temporarily, how investors are pricing the next phase of the AI buildout.

Meta’s Muse AI Agent Ignites a Chip Stock Rally, Pushing AMD Past $1 Trillion

What Happened?