Meta Launches Enterprise AI Platform, Hires MongoDB CEO to Lead Business Push
TL;DR
- Meta has created a dedicated Enterprise AI division to sell Muse, Meta Business Agent, Muse API and Muse Code directly to businesses and developers.
- Former MongoDB CEO Dev Ittycheria has been hired to lead the push as Meta's new enterprise chief, reporting to CEO Mark Zuckerberg.
- The move puts Meta in direct competition with Microsoft, OpenAI, Google and Anthropic for enterprise AI budgets.
A MongoDB Veteran Takes The Helm
Meta is making its biggest enterprise push to date. The company confirmed Monday it has formed a new division focused solely on bringing artificial intelligence products to businesses, and hired former MongoDB CEO Dev Ittycheria to run it.
Ittycheria, who stepped down from MongoDB earlier this month after 11 years leading the database giant, will join Meta as Chief Enterprise Officer, reporting directly to Mark Zuckerberg. In a memo announcing the move, Zuckerberg said Ittycheria was brought on to build Meta's business AI effort from startup mode into a full-scale enterprise business.
The hire is a clear signal of intent. Ittycheria is widely credited with taking MongoDB from open-source startup to a $20 billion-plus enterprise software leader with more than 50,000 customers. His experience selling infrastructure, developer tools, and cloud services to CIOs is exactly what Meta has historically lacked.
What Meta Is Actually Selling
The new Enterprise AI division will house four core products under one go-to-market organization.
First is Muse for Business, a version of Meta's AI assistant tailored for workplace use with admin controls, data retention policies, integrations with Microsoft 365, Google Workspace, Slack and SAP, and deployment options including private cloud and on-premises for regulated industries.
Second is Meta Business Agent, the company's no-code agent builder for sales, customer support, marketing and commerce. Businesses can deploy agents across WhatsApp, Instagram, Facebook Messenger, websites and call centers. Meta says early testers including telecom, retail and financial services firms have used it to automate customer inquiries and boost conversion.
Third is the Muse API, which gives developers direct access to Meta's Llama and Muse models with enterprise SLAs, fine-tuning, retrieval-augmented generation tooling, and compliance certifications including SOC 2, ISO 27001 and HIPAA readiness.
Fourth is Muse Code, an AI coding assistant aimed at enterprise engineering teams, competing head-on with GitHub Copilot and Cursor. Meta says it is already used internally by thousands of its own engineers.
Why Now
The timing is no accident. Meta has spent billions building Llama open models and its Superintelligence Labs, but until now has made most of its money from consumer advertising. Enterprise AI, by contrast, is now a $200 billion-plus annual market growing at breakneck speed.
Zuckerberg has told investors he sees 2026 as the inflection point where AI agents move from experimentation to large-scale enterprise deployment. With open-weight Llama models gaining traction with developers but facing monetization questions, a dedicated enterprise sales and product organization gives Meta a direct path to revenue.
The company also has distribution leverage few rivals can match: more than 200 million businesses already use WhatsApp Business, Instagram and Facebook for customer communication, giving Meta Business Agent a built-in channel.
Taking On Microsoft, OpenAI, Google and Anthropic
Meta's entry supercharges an already brutal enterprise AI race.
Microsoft and OpenAI dominate today with Azure OpenAI and Microsoft 365 Copilot entrenched in the Fortune 500. Google is pushing Gemini and Vertex AI, while Anthropic has won enterprise trust with Claude for coding and compliance-heavy industries. Salesforce, ServiceNow and Oracle are also embedding agents across workflows.
Meta's pitch is different on three fronts: open models that avoid vendor lock-in, lower cost per token at scale, and consumer-grade distribution via WhatsApp and Instagram where customers already are. Analysts say the MongoDB playbook of developer-first adoption followed by enterprise upsell could work well for the Muse API and Muse Code.
The challenges are real, however. Meta lacks a traditional enterprise sales force, faces lingering trust questions around data privacy, and will need to prove it can meet uptime, security and support expectations of large CIOs.
What To Watch Next
Meta says Ittycheria will begin building out a dedicated enterprise go-to-market team in the coming months, including sales, solutions engineering and partnerships, with headquarters in New York and San Francisco. The company is expected to detail pricing, expanded integrations and new Llama-powered agent features at its upcoming developer and business messaging events.
If successful, the division could transform Meta from an advertising company dabbling in enterprise into a true full-stack AI provider spanning models, APIs, agents and productivity tools.
Get All The Latest Updates Delivered Straight To Your Inbox For Free!