Meta's Muse hits 10M downloads in 40 days, beating AI rivals on retention

Meta's AI companion Muse has crossed 10 million U.S. downloads in just 40 days, a pace that puts it in rarefied air: only 11 other apps have hit that mark that quickly, and one of them is Meta's own Threads. The new data, reported by TechCrunch, also shows Muse is keeping users engaged at higher rates than its AI rivals - a signal that Meta's bet on a standalone companion app is resonating beyond the hype cycle.

For a company that has spent the past two years playing catch-up in generative AI, the download figure is a validation of Meta's distribution muscle. But the retention stat matters more: AI apps have historically struggled to convert curiosity into habit. If Muse is holding users better than ChatGPT or Gemini, it suggests Meta has cracked something in product design or onboarding that others haven't. The milestone also puts Muse in the same early-adoption tier as Threads, which hit 10 million downloads in its first week but later saw engagement cool. Whether Muse follows that arc or sustains its momentum is the open question.

Muse's launch strategy is worth unpacking. Rather than burying the assistant inside Instagram or WhatsApp, Meta chose to ship a standalone app, giving it a direct line to app-store rankings and a clean testing ground for AI-native features. That move lets Meta compete head-to-head with OpenAI's ChatGPT, Google's Gemini, and a growing field of companion bots, while also collecting fresh user signals to feed its Llama model training. For a company that has historically monetized attention through social feeds, a standalone AI product represents a new kind of surface - one that could eventually carry subscriptions, ads, or both.

The retention advantage is the detail rivals will be studying. Most AI apps see a steep drop-off after the first week, as users realize the novelty wears off. If Muse is retaining at higher rates, it likely reflects Meta's ability to integrate the app with its existing social graph - friends, groups, and content that keep people coming back. It also suggests that Meta's AI assistant, which already appears across its family of apps, is benefiting from familiarity and trust that a brand-new chatbot from a startup cannot easily replicate.

Regulatory context matters here. Meta's AI ambitions have repeatedly collided with privacy rules, particularly in Europe, where the company has had to pause or adjust features over how user data is used for training. The company has already navigated similar battles with its Llama models, which are open-source but have faced restrictions in some regions. The U.S. download milestone is a bright spot, but the company's ability to scale Muse globally will depend on navigating those constraints. For executives watching from the sidelines, the lesson is that consumer AI adoption is not just a product race - it is also a regulatory and data-governance race, and Meta's home-market advantage in the U.S. gives it a head start that may not translate abroad.

The broader implication for the AI landscape is that distribution still beats raw model quality in the early innings. OpenAI and Google have arguably more advanced models, but Meta has the largest installed base of any consumer platform on Earth. Muse's 10 million downloads in 40 days is a reminder that when a company with Meta's reach decides to push a product, the app-store charts can move overnight. The real test is whether those users stay, and the early retention data suggests they are. And unlike the early days of Threads, which saw a surge of curiosity-driven signups followed by a plateau, Muse is showing stickiness that suggests the AI companion category has real legs.

For founders and operators building consumer AI, the takeaway is twofold. First, a standalone app can be a powerful wedge, but it needs a hook that ties back to an existing network or habit. Second, retention is the metric that will separate winners from also-rans in the AI companion category. Meta has shown it can get users in the door; now the industry will be watching to see if Muse can keep them - and what that means for the next wave of AI products.