Meta has announced a dedicated enterprise AI business, though it has not announced a complete suite that companies can immediately buy and deploy.
In its September 28 Meta Enterprise Platform announcement, the company named four areas of initial focus: the Muse agent, Meta Business Agent, Muse API and Muse Code. CEO Mark Zuckerberg called the platform the “next major pillar” of Meta’s business. His new enterprise chief described the goal as turning Meta’s AI stack into products and services companies can deploy for themselves.
The move marks a change in direction for a company whose AI strategy has largely reached people through consumer products and businesses through advertising tools. For now, it is an early statement of intent. Meta did not publish platform pricing, a general-availability schedule, supported regions, launch customers or a detailed deployment architecture. Buyers still lack the details needed to know what they can purchase under a contract today.
Meta Named Four Products, but Few Enterprise Terms
Meta presents the platform as a way to bring together existing AI work for businesses and developers. Its initial focus spans several kinds of product:
- Muse is Meta’s personal AI agent. Meta says it can carry out tasks on a person’s behalf; the enterprise announcement does not specify how it would be configured for an organization.
- Meta Business Agent points toward AI interactions between businesses and their customers. Meta had already begun pursuing business-facing agents before this platform announcement.
- Muse API signals an offering for developers, but Meta’s announcement does not spell out its interface, commercial terms or rollout.
- Muse Code is Meta’s coding agent. The company has discussed it separately as a tool for longer engineering tasks.
Together, they suggest a broader pitch than supplying an AI model for someone else’s application. Meta wants a place in customer interactions, employee or individual workflows, developer integrations and software engineering.
The announcement does not specify four new enterprise plans. There is no feature matrix explaining which products are sold together, which can be bought separately, or what administrative controls come with each one. Meta calls them an “initial focus”; it has not confirmed that every component is available through the platform on identical terms.
Existing Muse Products Do Not Set the Platform’s Availability
Some of the named technology predates the enterprise announcement. Meta introduced Muse as a personal AI agent earlier in September, describing a product that can work across apps, use a browser and request approval before sensitive actions. At that announcement, Meta said Muse was rolling out in the United States on iOS, Android and muse.ai. Those details establish the consumer rollout, not the geographic availability of an enterprise contract.
Meta’s Connect developer recap said Muse Code was out of beta and available on Windows. It also described Meta Model API as generally available globally, including support for enterprise workloads. Those statements are more concrete than the enterprise platform announcement, but the names require care: Meta did not explain in its September 28 post how the named Muse API relates to Meta Model API, or whether their availability and purchasing terms are the same.
A developer’s ability to use a Meta API, or someone’s ability to use Muse Code, does not answer how a large company would procure, administer or deploy the newly announced platform. Meta has connected these products under an enterprise strategy without publishing how that strategy will work for buyers.
Zuckerberg Is Building a Business Beyond Advertising
Meta already works with advertisers and businesses that use its apps to reach customers. In the platform announcement, Zuckerberg pointed to those relationships, along with Meta’s models, agents and infrastructure, as reasons the company thinks it can serve enterprise buyers.
Selling AI software directly asks something different of Meta. An advertiser paying to reach an audience is not necessarily buying a coding agent for engineers or approving access to company data for an AI assistant. Those purchases involve different technical evaluations, procurement processes and support expectations.
The direction was visible before September. During Meta’s July earnings discussion, Zuckerberg described opportunities to sell businesses APIs, agents and other services, according to TechCrunch’s account of his comments. He also acknowledged that enterprise sales would require a “different muscle” from Meta’s historical business. The new platform gives that ambition an organizational home.
Chirantan “CJ” Desai will lead it as Chief Enterprise Platform Officer, reporting directly to Zuckerberg. Meta says he joins from MongoDB, where he was president and CEO, after senior roles at Cloudflare and ServiceNow. The appointment and reporting line show a dedicated effort, though they say nothing yet about how much revenue it can generate. The Wall Street Journal’s account framed the push as an attempt to earn a return from Meta’s AI investment by selling more tools and services to companies.
Buyers Still Need Prices, Controls and Deployment Answers
Meta has not disclosed prices or billing units, general-availability dates, supported enterprise regions, or named launch customers. It also has not explained in detail where the enterprise products run, how they connect to company systems or what deployment options buyers can choose.
The unanswered questions extend beyond cost. An API customer needs to understand usage limits and support. A security team assessing an agent that acts across applications needs to know how access is granted, monitored and revoked. An organization considering a coding agent needs to assess what code the tool can access and what controls apply to its output. Meta’s announcement neither establishes nor rules out those features.
Meta says security and privacy are being built into its enterprise products from the outset. It has described specific protections for consumer Muse, including a dedicated virtual machine and controls over connected services. The enterprise announcement does not provide enough detail to determine whether consumer Muse’s architecture, data handling or contractual protections apply to every product sold through the platform.
Frequently Asked Questions
4 questions
1What Is Meta Enterprise Platform?
Meta Enterprise Platform is Meta’s newly announced effort to turn its AI stack into products and services for companies. Its initial focus includes the Muse agent, Meta Business Agent, Muse API and Muse Code. Meta describes it as a new business pillar, but its announcement does not establish that all four products are available together as a complete enterprise suite.
2
Sources
- Meta Enterprise Platform announcementabout.fb.com
- Muse as a personal AI agentabout.fb.com
- Connect developer recapdevelopers.meta.com
- TechCrunch’s account of his commentstechcrunch.com
- Wall Street Journal’s accountwsj.com
