Meta Enterprise AI: New Revenue Streams Beyond Advertising

Meta's Enterprise AI Expansion

In June, Meta entered the enterprise AI market with a new AI agent aimed at businesses, targeting customer service, support, and daily operations. But the tech giant's enterprise AI ambitions are far more expansive, as Meta CEO Mark Zuckerberg detailed on the company's second-quarter earnings call.

Zuckerberg stated, “We see a large enterprise opportunity to sell to businesses, including APIs, business agents, potentially selling compute directly, and other services that we’re building for large customers.” These additions could create new revenue streams beyond advertising—which currently drives the bulk of Meta's business—and subscriptions, which contribute a smaller share.

New Revenue Opportunities from AI Agents

Initially, Meta will focus on serving its existing advertiser base by offering AI agents that work across messaging apps and other platforms. These agents allow businesses to interact with customers through an AI interface.

Performance-Based Pricing Model

Zuckerberg explained the business model: “And, just like the ad system, effectively, we will get paid when we deliver results for those businesses.” He views this as an extension of Meta's existing sales and partnerships with millions of advertisers and hundreds of millions of small businesses.

Internal Tools for External Customers

Meta also plans to expand beyond small businesses by offering its internal productivity tools to external customers. Zuckerberg noted, “We’re building coding and developing and internal productivity tools partially because we need to build them ourselves. Now that we have those, we feel like there’s a large opportunity to serve—whether that’s small businesses or larger businesses.”

However, this shift won't be easy. Zuckerberg admitted that selling to the enterprise requires “a different muscle” than Meta has historically flexed.

Selling Compute and Balancing Priorities

Meta is also considering selling compute power to enterprise customers. The company emphasized it has the opportunity to sell compute “at a significant premium over what we paid for it.”

A Portfolio Approach to Compute

Zuckerberg cautioned investors against selling all compute for short-term profit. Instead, Meta views its compute infrastructure as a portfolio with a mix of long-term and short-term plans. He stated, “As we get closer to personal superintelligence, we are going to need hardware that allows you to seamlessly interact with it.”

Agentic AI for Businesses and Consumers

The earnings call also highlighted Meta's ambitions around agentic AI—systems that can act on a person's or business's behalf, rather than just answering questions.

Consumer AI and Smart Glasses

Consumers are promised “personal AI agents” as well as AI smart glasses that can interact with the world in front of them. Meta is using large language models to rapidly build out its social app suite.

Recent Launches and Future Plans

Recent launches include an app for Marketplace sellers, one for Facebook Groups, a vibe-coded games app, and other experiments. Zuckerberg teased more on the way, stating, “I expect it to become a lot easier to ship new apps. So we are planning to build out more ideas and use our recommendation systems to scale them.”