If the download metrics are the sole measure, Muse has already secured victory. Yet Mark Zuckerberg's ambitions clearly extend beyond mere numbers.

At the annual Connect developer conference held in Menlo Park, California on September 23, the message was unmistakable: Muse, a personal AI agent launched just two weeks prior, is destined to become the core intelligence driving every future Meta hardware product. From lightweight VR headsets to camera-free smart glasses, and even a miniature terminal that clips onto a keychain, Muse is integrated into every new offering. This marks a rare moment of perfect synchronization between product development and market timing for Meta in the AI arena.

Over the past two years, Meta's position in the AI landscape was awkward—significant capital and computing power were deployed, yet a consumer-friendly product remained elusive. The arrival of Muse has shifted this dynamic.

Weighing 100 Grams and Priced at $1,299, Meta Redefines the VR Headset

The hardware generating the most buzz at the conference is the Meta VR Glasses. The defining feature isn't its functionality but its weight—approximately 100 grams, roughly one-sixth that of Apple Inc.'s Vision Pro and merely one-fifth of Meta's own Quest 3S. This breakthrough is achieved through a radical architectural overhaul. Meta has externalized all computational elements—processor, battery, and cooling fan—into a pocket-sized standalone puck, leaving the glasses with only the display and sensor modules. This design trade-off results in a slight reduction in display resolution at approximately 2.5K per eye, lower than the Vision Pro's 4K, but it delivers a superior three-hour battery life compared to the Vision Pro, with support for continuous plugged-in use.

The $1,299 price point, roughly one-third of the Vision Pro's $3,699 tag, reflects a clear pricing strategy: Meta isn't seeking a head-on battle with Apple in the premium segment. Instead, it aims for an 'affordable premium' positioning to attract consumers interested in VR but hesitant to spend nearly four thousand dollars on an early adopter device. Powered by Qualcomm's Snapdragon Reality Elite processor, the headset supports eye-tracking and hand-gesture controls, allowing users to interact directly with the integrated Muse assistant to complete tasks. The product is slated for a spring 2027 release.

Smart Glasses Without a Camera: Meta's Sincerity or Calculated Move?

Privacy controversies are an unavoidable obstacle for Meta's smart glasses. Over the past year, the product has been branded with an unfavorable label, facing bans in New York state courts, restrictions in Australian public pools and London bars, and reports of individuals using the glasses to capture images of women without consent for social media posting. In March, a 230-page class-action lawsuit was filed in California federal court, with over 70 consumers alleging Meta misled users regarding privacy protections. The lawsuit also brought to light that Meta's AI training annotators had reviewed private video footage uploaded by users, including changing-room scenes.

Against this backdrop, the launch of the Ray-Ban Meta Audio, Meta's first camera-free AI eyewear, can be interpreted as a direct response. Priced from $349, these glasses weigh 43 grams and offer up to 12 hours of single-charge battery life, extending to 60 hours with the charging case. Functionality centers on audio playback, calls, voice translation, and voice interaction with Muse. Designed by EssilorLuxottica, the product is available in 23 color and lens combinations, with pre-orders opening on October 13. Concurrently, the camera-equipped Ray-Ban Meta Gen 3 was also announced, starting at $449, featuring a 12-megapixel camera and a six-microphone array that Meta claims reduces background noise by over 90%.

Meta's CTO, Andrew Bosworth, has previously expressed a desire for a 'few bad actors' not to tarnish public perception of smart glasses, reiterating that Meta has implemented multiple privacy safeguards, including an automatic recording indicator light and a system that disables the camera if the indicator is obstructed. The release of the camera-less version offers consumers a 'no-trust-required' option at the product level—those who distrust the camera can directly purchase the alternative.

Charm: A Keychain-Sized AI Portal

Nearing the end of the presentation, Zuckerberg unveiled what became the most talked-about item of the evening—the Muse Charm. This palm-sized device features a 2-inch OLED touchscreen and a built-in 5G modem, running on a new operating system designed specifically for Muse. It comes equipped with a speaker, microphone, front and rear cameras, and a USB-C port, and supports fingerprint unlocking and media playback. Users can carry it via a wristband, lanyard, or simply in a pocket.

Charm's positioning isn't as a smartphone replacement but as a dedicated portal to Muse. Pressing the fingerprint sensor initiates direct AI conversation without needing to unlock a phone or open an app. Zuckerberg stated at the event, 'We've packed the full Muse experience—including real-time voice and digital avatars—into a device that fits on a keychain.' The device was designed by the lab led by Alan Dye, former head of interface design at Apple, and aesthetically resembles the 90s virtual pet device, the Tamagotchi. Its transparent casing exposes the internal circuitry, while the screen displays a customizable virtual Muse character whose appearance, outfit, and voice can be tailored. When two Charm users are in proximity, the devices can recognize and interact with each other.

Notably, Meta originally planned a 2027 launch, but Zuckerberg personally pushed the team to advance the timeline to before the end of 2026 to align with the holiday shopping season. However, Bosworth admitted that the Charm is more of an 'exploration' and a typical 'V0' product, anticipating that the number of users accessing Muse through phones and smart glasses will still significantly outnumber Charm users. The history of standalone AI hardware is not favorable—both the Humane AI Pin and Rabbit R1 attempted to decouple AI from smartphones as independent devices, with underwhelming market performance. Charm's significance may not lie in its own sales volume but in the signal it sends to developers and the market: Muse's form factor is not confined to a smartphone screen, and Meta is open to exploring any viable hardware carrier for it.

Muse's Foundation: From 2.5 Million Downloads to a 36 Billion User Base

The viability of Connect's hardware offensive hinges on Muse's proven traction. Since its launch on September 8, Muse's download figures have consistently exceeded market expectations. According to Sensor Tower data, cumulative downloads reached approximately 2.6 million in the first 13 days, with 642,000 US mobile daily active users—significantly higher than ChatGPT's 231,000 at its equivalent launch period. Meta's stock surged 11.4% on the 13th day post-Muse launch, adding nearly $200 billion to its market capitalization.

However, what excites Wall Street isn't the download numbers themselves but Muse's distribution logic. Over 95% of Muse users are also Facebook users, and 63% are Instagram users. This means Meta doesn't need to acquire users from scratch—it possesses a social empire with 3.6 billion daily active users, and Muse simply needs to convert within this existing pool. The monetization path is also emerging. Muse currently offers a free tier with a weekly quota of 100 million tokens, with paid plans at $20 per month for 500 million tokens and $100 per month for 3 billion tokens.

Analysts at Jefferies project that if Muse reaches 1 billion users by the end of 2027 with a 3% conversion to paid, annualized revenue could reach $10.8 billion. Evercore's Mark Mahaney offers a more straightforward estimate: if just 1% of Meta's 3.6 billion users pay $20 monthly, it would generate roughly $8 billion in annual revenue. On the e-commerce front, Meta is accelerating its efforts. Muse has integrated with retailers including Walmart, Best Buy, Gap, Sephora, and Wayfair, as well as the PayPal payment system, and has added Expedia for travel bookings and Instacart for grocery shopping. Zuckerberg has clearly stated Meta's business model is to 'take a small fee from transactions.' Interestingly, Amazon has blocked Muse from browsing its store, which paradoxically highlights the strategic value of partners like Walmart.

Muse will also support Mac operations, allowing users to delegate tasks such as file management, email handling, and schedule organization to the AI agent, which continues executing queued tasks even after the user steps away from the computer. Bain & Company forecasts that the US agentic e-commerce market will reach $300 billion to $500 billion by 2030, representing 15% to 25% of the entire US e-commerce market. Meta is clearly positioning itself to capture this market.

Strategic Synthesis: Muse as the Hub Connecting Hardware and Ecosystem

Connecting the dots across Connect's product announcements, Meta's strategic logic is coherent: Muse serves as the central hub, hardware provides the entry points, and the social empire's user base acts as the distribution channel. The VR glasses target immersive entertainment scenarios, the audio glasses cater to daily commutes and light interactions, and the Charm addresses on-the-go rapid access. Each scenario fulfills a distinct role, with the collective goal of making Muse the most frequent AI touchpoint in users' daily lives.

This path is not without challenges. Meta's capital expenditure guidance for 2026 is a substantial $130 billion to $145 billion, primarily directed at data centers and AI infrastructure. At this investment scale, market expectations for Muse have elevated beyond being merely a useful AI product—it is now viewed as a potential growth engine to sustain Meta's next decade. Privacy controversies will also not dissipate in the short term. However, the product cadence and strategic clarity displayed at this conference at least present investors with a tangible possibility: at the inflection point where AI transitions from 'conversational' to 'operational,' Meta has not fallen behind.