Meta shareholders have spent much of 2026 oscillating between enthusiasm for the company’s artificial intelligence ambitions and concern over the enormous cost of pursuing them. Earnings reports have repeatedly produced sharp moves in the stock, while investors have struggled to settle on whether Meta’s escalating AI spending represents a competitive advantage or a growing financial burden.
The latest shift has been decisively more optimistic.
Meta’s shares have surged following the rapid adoption of Muse, the company’s personal AI agent launched on September 8. The reaction suggests that investors are beginning to see something different from another promise of what AI could eventually become for Meta. Muse offers an early example of a consumer product that could potentially turn the company’s enormous investment in AI into a new source of revenue and user engagement.
Register for the next Tekedia Mini-MBA.
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Meta shares were up 18% for the year through Thursday, but that figure masks considerable volatility. The company, valued at nearly $2 trillion, has experienced the sort of sharp sentiment swings more commonly associated with smaller technology companies.
The underlying debate has remained consistent. Meta has been spending heavily on AI infrastructure, models, data centers, talent, and devices while investors have sought evidence that those investments will eventually generate returns beyond improving advertising.
Muse is providing the clearest test yet.
From AI assistant to AI agent
The significance of Muse lies partly in the broader transition taking place across the AI industry.
Early consumer AI products largely revolved around answering questions, generating text and images, summarizing information and responding to prompts. AI agents promise something more consequential: systems that can actually perform tasks on a user’s behalf. That could include managing email, completing online forms, navigating websites and coordinating activities across different digital services.
For Meta, the opportunity is large because the company already controls a massive consumer distribution network through Facebook, Instagram, WhatsApp and its other platforms.
Andrew Boone, an analyst covering Meta at Citizens, described Muse as potentially representing another major transition in consumer AI.
“With Muse appearing to reach the next AI inflection point, the same way Claude Code did for enterprise AI last winter, and ChatGPT did originally with Chatbots, Meta is increasingly well positioned to grow share of consumer AI usage,” Boone wrote in a Thursday report.
Meta does not need to create an audience for AI from scratch. It already has billions of people using its services. The challenge is converting some of that existing audience into regular users of its AI products. That gives Meta a distribution advantage that many AI startups do not possess.
The company can introduce AI capabilities to people who are already inside its ecosystem rather than spending enormous sums acquiring users independently. If Muse becomes a frequently used personal agent, Meta could potentially use that engagement to develop subscription products, commerce services, advertising opportunities, or other forms of monetization.
The company is also extending the strategy beyond software.
Meta’s AI push includes consumer devices such as its Charm dongle and $1,300 VR Glasses. The broader ambition is to make AI accessible through devices that could eventually reduce the importance of the traditional smartphone interface.
Instead of opening an application and tapping through menus, users could increasingly interact with digital services through conversational AI and wearable devices. That is a much larger proposition than simply building another chatbot.
Investors Are Pricing in The Possibility
The market’s response shows how quickly expectations can change when investors see a potential commercial application for AI. JPMorgan analysts raised their price target for Meta by $100, while Raymond James increased its target by $210. Those moves came as Meta’s stock was already trading at a substantial valuation.
The reaction highlights the central issue surrounding Meta’s AI spending. Investors have not necessarily objected to the company spending heavily on AI. The uncertainty has been whether those investments will produce economic returns large enough to justify their scale.
Meta has the financial resources to sustain the spending. Its advertising business generates enormous cash flow, giving Zuckerberg considerable room to invest in infrastructure and research while continuing to develop new products.
But the market has increasingly wanted evidence of where the payoff could come from.
Muse provides a more tangible answer than a distant promise about artificial general intelligence or future AI capabilities. But that does not mean the business model has been established.
The application is still extremely new, and early download numbers or app-store rankings do not establish long-term retention. Consumer technology has repeatedly produced products that attract enormous attention immediately after launch and then struggle to maintain that momentum.
That has resulted in a notable contrast between adoption and durable usage.
For Muse to justify the market’s enthusiasm, users will ultimately need to return frequently and allow the product to become embedded in everyday activities. Meta would then need to develop a sustainable way of monetizing that engagement without undermining user trust.
The last part could prove particularly difficult because personal AI agents potentially have access to much more sensitive information than conventional social media applications.
An agent capable of managing email, interacting with websites, making purchases, or organizing personal tasks could have access to financial information, communications, preferences, and other data. The commercial opportunity is therefore accompanied by considerably higher privacy and cybersecurity requirements.
Zuckerberg’s Bigger Bet
Zuckerberg’s ambition goes beyond building a successful AI application. Meta is effectively betting that AI could become the next major consumer computing platform.
Analysts believe that is why the company’s investments in AI models, agents, smart glasses, virtual reality, and other devices are increasingly connected. The underlying strategy is to control more of the interface between consumers and digital services.
Apple established enormous influence by controlling the operating system and hardware layer through the iPhone and iOS. Meta cannot realistically displace that ecosystem directly. Instead, it is pursuing a different possibility: that consumers may eventually interact with technology less through conventional screens and applications and more through AI assistants and wearable devices.
If that transition occurs, the company with the strongest consumer AI assistant and the largest installed audience could have a significant distribution advantage.
But there are several unresolved questions.
Consumers may continue to prefer traditional applications and smartphones. AI agents may prove useful for specific tasks without becoming a primary computing interface. Competitors including OpenAI, Google, Anthropic and others are also developing increasingly capable agents.
And Meta’s existing advantage in social media does not automatically translate into leadership in AI.
The company therefore faces the same problem confronting much of the technology industry: converting technical capability into durable consumer behavior and, eventually, profits.
For now, investors appear willing to give Meta more credit for that possibility.
The latest rally is less about the immediate financial contribution from Muse than what the product represents. After years of spending on AI infrastructure, models and hardware, Meta has produced a consumer-facing application that gives investors a clearer picture of how those investments might eventually become a business.
That makes Muse an important test, but not yet proof, of Meta’s AI strategy.
The company’s next challenge will be turning initial enthusiasm into sustained usage, then sustained usage into revenue. Analysts predict that if it manages to do that while leveraging its enormous existing audience and expanding into AI-enabled devices, Meta could establish a meaningful position in the emerging agent economy.
But if users move on after the initial excitement, the market will have to return to the question that has followed Meta throughout 2026: how much should investors pay today for an AI future that remains largely dependent on execution?