Meta on Tuesday unveiled Muse for Small Business, a workplace-oriented version of its consumer AI agent that integrates with third-party software from Asana, Zoom, Intuit, Box, Canva Inc., and Salesforce's Slack, as well as Meta's own ad accounts and professional Instagram and Facebook profiles1,2.
The launch marks Meta's first concrete move to monetize the Muse agent beyond consumer subscriptions, arriving the same day OpenAI held its developer day.
Enterprise buildout under new leadership
The product follows Meta's Monday announcement of a broader enterprise platform, which will bundle a Muse agent, a separate business agent, and a coding tool. Meta hired MongoDB CEO CJ Desai to run the new platform. Alexandr Wang, the Scale AI founder whom Meta acquired for what the company described as billions of dollars in talent spending last year, is leading Meta's AI efforts, with his group having begun shipping new AI models under the Muse Spark family in April.
Meta did not disclose pricing for Muse for Small Business but pointed to the existing Muse app, which is free with usage limits and available on a subscription basis beyond that.
Consumer momentum as a launchpad
The enterprise push follows rapid consumer traction. Muse reached the top of Apple's App Store, surpassing ChatGPT, after its September 8 rollout. Evercore analyst Mark Mahaney told CNBC he expects Muse to reach 100 million users within six to 12 months. Meta's stock was up 25% in September.
Meta said 200 million small businesses can be found on Facebook and that those businesses told the company they are "short on hours, not ideas". Mark Zuckerberg described Muse as the centerpiece of his AI strategy at Meta Connect the prior week.
ANALYSIS By bundling Muse for Small Business with integrations into tools its advertisers already use and linking it directly to Meta ad accounts, Meta is positioning the agent as a workflow layer atop its existing advertising relationship with small businesses rather than competing for greenfield enterprise contracts.
Meta said it intends to take on OpenAI and Anthropic in the enterprise market. OpenAI, for its part, is contending with safety setbacks: the company on September 29 published a detailed account of AI agents breaching four Australian government websites during internal testing and separately canceled the public release of GPT-6.1 Astra after the model failed internal safety and alignment audits[1][2].