SAN FRANCISCO — Meta Platforms has officially confirmed what market researchers have been whispering for weeks: its breakthrough artificial intelligence application, Muse, has crossed the formidable threshold of 10 million active users. The announcement, delivered by CEO Mark Zuckerberg during Meta’s third-quarter earnings call on Wednesday, validates earlier data from market intelligence firm Sensor Tower, which estimated the milestone following a meteoric 40-day growth sprint.
Yet, beyond the initial download velocity, it is user engagement, heavy-duty agentic commerce, and long-term monetization metrics that have captured Wall Street’s attention. As Meta transitions Muse from its explosive introductory phase into a broader infrastructural pivot, the tech giant is setting its sights on redefining how consumers shop, communicate, and interact with hardware.
Main Facts: The Rise of Muse and Agentic Commerce
The core narrative of Meta’s latest earnings disclosure centers on Muse’s rapid ascent and its surprisingly lucrative utility. According to Zuckerberg, roughly 25% of all users who set up Muse return to engage with the application on a daily basis, a retention rate that signifies robust product-market fit.
However, the most startling revelation from the earnings call was the sheer purchasing power exhibited by users conducting transactions through the AI agent. More than a quarter of Muse’s engaged shoppers—those utilizing the AI to navigate purchases—recorded an average basket size exceeding $200. This high-ticket volume has taken even industry veterans by surprise.
Muse achieves these seamless transactions through deep integration with Plaid, the financial technology leader that securely connects consumer payment applications to personal bank accounts. Zuckerberg noted that Plaid executives reported unprecedented activity volumes through their connector, stating they had never witnessed transactional velocity and engagement of this magnitude on any other service.
Beyond commerce, Muse has quickly established itself as a versatile personal productivity tool. Users are increasingly linking the AI agent to sensitive personal data ecosystems, including email clients and calendars. Zuckerberg emphasized that this widespread adoption for high-trust tasks underscores consumer confidence in Meta’s security architecture.
To safeguard privacy, Muse utilizes a proprietary security setup dubbed the "Muse Secure VM." Furthermore, Meta has previewed an upcoming "Muse Confidential VM" iteration designed with zero-knowledge architecture, ensuring that even Meta itself cannot view or inspect user content.
Chronology: From Concept to 10-Million-User Milestone
The trajectory of Muse highlights the hyper-accelerated timelines governing the current generative AI landscape:
- Late 2025 / Early 2026: Development and internal testing of Meta’s next-generation multimodal agentic AI application culminate in a targeted public beta and official rollout.
- August – October 2026: Muse experiences an unprecedented viral adoption wave, spreading rapidly across domestic app stores.
- Mid-October 2026: Market intelligence firm Sensor Tower releases independent tracking data confirming that Muse successfully hit 10 million U.S. downloads in just 40 days—a rare feat achieved by only a tiny fraction of software applications in history.
- Wednesday (Earnings Call): Mark Zuckerberg and Meta executive leadership officially confirm the 10 million user milestone during the Q3 earnings call, detailing heavy monetization potential, security protocols, and hardware integration strategies.
Supporting Data and Financial Projections
Meta’s financial outlook for Muse shatters conventional expectations for consumer AI software. While consumer-facing AI tools have historically struggled to build sustainable, high-margin revenue models without exorbitant subscription fees, Meta is charting a hybrid course.
Monetization Model
Meta intends to keep Muse fundamentally free for the vast majority of baseline users. Monetization will primarily occur through two distinct channels:
- Commerce Commissions: Meta will retain a percentage fee on transactions processed through Muse’s agentic commerce network.
- Power User Tiers: Advanced capabilities—such as access to higher-tier, computationally demanding AI models or expanded token allotments—will be walled behind a paid subscription model.
Revenue Expectations per User
Most strikingly, Meta’s internal forecasts project that Muse will eventually generate more revenue per user than its core "family of apps" (Facebook, Instagram, WhatsApp, and Messenger) achieve today. For context, Zuckerberg noted that the current family of apps yields "several hundreds of dollars annually" per user through targeted advertising and commerce pipelines. Surpassing this baseline via an AI agent represents a seismic shift in Meta’s top-line potential.
Compute Costs and Profitability Timeline
Running state-of-the-art agentic models is notoriously expensive. Acknowledging near-term financial friction, Zuckerberg noted that Muse is currently "very compute-intensive to run." To counteract this, Meta is aggressively building out data center infrastructure capable of servicing over a billion people in the coming years.
Simultaneously, engineering optimizations are expected to drive down compute costs by more than 10x over the next year. Once these efficiency gains are realized and transactional volume scales, Meta expects Muse to operate as a high-margin, profitable entity.
Official Responses and Strategic Implications
The success of Muse does not exist in a vacuum; it serves as a critical linchpin for Meta’s broader hardware, software, and corporate philosophy.
Reality Labs and Smart Glasses Integration
For years, investors have scrutinized Meta’s Reality Labs division—the R&D powerhouse responsible for virtual reality, the metaverse, and smart glasses—due to its staggering financial burn rate. While Reality Labs continues to lose billions, the bleeding is slowing. Financial projections indicate that division losses will shrink from $19.1 billion in 2025 to $15 billion this year.
Muse is slated to become the cognitive core of Meta’s expanding smart glasses ecosystem, providing an always-on, multimodal assistant that bridges the gap between digital processing and real-world perception.
The "Muse Network": Moving Beyond Single-Player AI
Looking forward, Zuckerberg teased the conceptualization of the "Muse Network." Describing current AI interactions as a "single-player game," Zuckerberg outlined a future where individual AI agents can seamlessly collaborate with the Muses of friends and family members.
In this interconnected framework, commercial transactions will evolve past traditional web browsing. Consumers will no longer navigate to static business websites; instead, their personal Muses will negotiate, transact, and communicate directly with corporate business agents.
Financial and Industry Implications
The implications of Meta’s Muse strategy extend far beyond social media and consumer hardware:
- Disruption of Traditional E-commerce: If agentic shopping with average basket sizes over $200 becomes the default consumer habit, retail brands will be forced to optimize their backend systems for AI-to-AI negotiations rather than human-centric search engine optimization (SEO).
- Fintech Partnerships: The validation of Plaid’s connector by massive consumer volumes cements financial APIs as essential infrastructure for generative AI agents.
- The AI Profitability Race: As tech giants grapple with the crushing capital expenditures of AI data centers, Meta’s roadmap of combining free-tier utility with transaction-based monetization and dramatic compute efficiency gains offers a blueprint for sustainable AI profitability.
As Meta scales its compute capacity and prepares to roll out advanced privacy features like the Confidential VM, Muse stands as an early harbinger of the post-app era—where intelligent agents act not just as search tools, but as autonomous proxies for everyday human commerce, productivity, and connection.
