Meta Launches Enterprise AI Division to Monetize $600 Billion Infrastructure Buildout

Meta is formally entering the enterprise software market, establishing a dedicated business unit to sell artificial intelligence tools and services directly to corporations. The move marks the company’s most aggressive effort to build a revenue stream independent of its advertising-dependent model on Facebook and Instagram.

Mark Zuckerberg announced the creation of the Meta Enterprise Platform in a Facebook post, describing the initiative as the “next major pillar of our business.” To lead the division, Meta hired Chirantan “CJ” Desai as Chief Enterprise Platform Officer. Desai previously served as CEO of MongoDB, the publicly traded database software company, bringing direct experience managing enterprise sales cycles and developer-focused platform businesses.

The new unit will commercialize several products Meta has developed over recent months. These include Muse, an AI agent framework; Meta Business Agent, designed for corporate workflow automation; Muse API, which allows developers to integrate Meta’s models into third-party applications; and Muse Code, a programming assistance tool.

By packaging these offerings for corporate buyers, Meta is stepping directly into territory dominated by Microsoft, Salesforce, and ServiceNow. Microsoft has integrated Copilot across its Office and Azure ecosystems, while Salesforce has embedded its Einstein AI into customer relationship management workflows. Unlike those competitors, which built enterprise relationships over decades of selling business software, Meta is attempting to leverage its massive consumer data infrastructure and proprietary large language models — primarily its open-weight Llama family — to attract corporate clients who want to run AI workloads without relying entirely on closed-source providers.

The pivot arrives as Meta faces mounting scrutiny over the return on investment from its unprecedented capital expenditure. Zuckerberg has stated the company expects to spend roughly $600 billion on AI development over the next two years, with the majority directed toward constructing new data centers and acquiring specialized hardware such as Nvidia GPUs.

That spending trajectory has pressured Meta to demonstrate how it intends to generate returns beyond its core ad business. Consumer-facing AI subscriptions introduced recently for individual users and small business owners have not yet proven capable of scaling into a reliable revenue driver. Enterprise contracts, which typically carry higher margins and longer retention rates than consumer subscriptions, offer a more viable path to justifying the capital outlay.

Beyond software licensing, Meta has explored monetizing its physical infrastructure. Industry reports in recent months indicated the company considered leasing excess data center capacity to outside organizations, a practice pioneered by Amazon Web Services that transformed Amazon’s internal server costs into its most profitable division. Meta has not confirmed specific plans to launch a cloud hosting service, but the enterprise platform’s reliance on “large-scale infrastructure” suggests its compute resources will play a central role in whatever services are ultimately offered.

Desai’s appointment signals Meta intends to approach corporate sales with traditional enterprise rigor rather than treating business tools as secondary features appended to its social networks. His tenure at MongoDB involved scaling the company’s transition from open-source database distribution to a fully managed enterprise cloud service, a playbook closely aligned with Meta’s current challenge of converting open-weight AI research into paid corporate deployments.