Meta's Muse Joins the AI Agent Race
AI-generated image · Bay Street Wire
The launch of Muse marks a shift toward proactive personal assistants, following a trend sparked by the open-source OpenClaw platform.
Meta has entered the AI agent market with the release of Muse, a consumer-facing tool that, as The Verge first reported, has reached 600,000 daily active users in the U.S. by an Apptopia estimate and recently topped App Store charts.
The emergence of Muse is part of a broader industry pivot toward agentic AI, largely inspired by OpenClaw. According to The Verge, OpenClaw was a one-man project created by Peter Steinberger that allowed users to interact with AI via messaging apps like Discord, Slack, and WhatsApp. The platform's success led to significant industry movement: OpenAI hired Steinberger in February, Google announced agent bets in May, and Apple committed to agents in June.
While Meta's Nat Friedman, head of product for Meta’s Superintelligence Labs, stated that Muse was built "from scratch," he admitted the product was "heavily inspired" by OpenClaw. This inspiration is evident in Muse's design and the use of identical file names and personality guidelines. Friedman noted that after using OpenClaw in January, he purchased hundreds of Mac Minis for his team; with Muse, he aimed to build a similar platform that could be made safe, secure, and scalable.
This shift toward proactive assistance is further evidenced by the AI agent platform Instinct. The Verge reports that Instinct users are employing agents for complex personal tasks, such as organizing bachelor parties, paying toll bills, and booking DMV appointments. Instinct's creator is currently fundraising at a $2.5 billion valuation.
Security remains a primary differentiator for the Big Tech versions. The Verge reports that 15 percent of OpenClaw’s skill repository contained "malicious instructions." In response, Meta CEO Mark Zuckerberg stated that Muse was built from the ground up specifically for security and privacy.