Meta sees much more potential in enterprise AI than just chatbots for customer service. During the presentation of the quarterly results, CEO Mark Zuckerberg (photo) outlined a strategy to sell enterprise AI services, computing power, and development tools. This is intended to create a new revenue stream alongside advertising, which still accounts for the largest share of Meta’s revenue.
In June, Meta took its first step into the enterprise AI market with an AI agent for businesses. Among other things, it supports organizations with customer interactions and day-to-day operational processes. According to Zuckerberg, however, that is just the beginning, TechCrunch reports.
“We see a large enterprise market where we can sell products and services to businesses, including APIs, AI agents, and possibly even direct sales of computing power and other services we’re developing for large customers,” he told investors.
AI agents as an extension of the advertising model
The initial target audience consists of the millions of businesses that already advertise on Facebook, Instagram, and WhatsApp. Meta wants to offer them AI agents that can independently handle customer inquiries via the companies’ own messaging platforms.
According to Zuckerberg, this aligns seamlessly with the existing revenue model. “Just like with our advertising system, we’re essentially paid when we deliver results for those businesses.” He described the new AI services as an extension of existing relationships with millions of advertisers and hundreds of millions of small businesses that already use Meta’s platforms.
Ultimately, Meta does not want to limit itself to smaller businesses. The company is also exploring how to commercially offer its internal software for software development and productivity to external customers. “Now that we’ve built these tools ourselves, we see a major opportunity to serve both small and large businesses with them,” Zuckerberg said.
He did acknowledge, however, that this market requires a different approach. According to him, selling to large enterprises is “a completely different discipline” from the advertising model on which Meta has built its success.
Computing power as a commercial product
In addition to software, Meta also sees opportunities to market AI infrastructure. The company is investing tens of billions of dollars in data centers and AI hardware, giving it significant computing capacity. According to Meta, that capacity can currently be sold to external customers at an attractive margin.
Nevertheless, the company intends to proceed cautiously in this regard. Zuckerberg warned that it would be “foolish” to sell all available computing power for short-term profit. Meta wants to retain sufficient capacity to develop its own AI models and future services.
A key part of the strategy is the development of so-called agentic AI: systems that can independently perform tasks on behalf of a user or organization, rather than simply answering questions.
This technology isn’t being developed exclusively for businesses. Meta is also working on personal AI agents for consumers and on smart AI glasses that can interpret their surroundings and respond accordingly.
In addition, the company is using large language models to develop new applications more quickly. Zuckerberg expects AI to significantly shorten the time to develop new products. “I expect it will become much easier to release new apps,” he said. As a result, Meta wants to experiment more with new applications and quickly bring them to the attention of the right users through its recommendation systems.