Meta Targets Enterprise AI Beyond Advertising

Meta (formerly Facebook) CEO Mark Zuckerberg stated during the company’s Q2 earnings call on July 29, 2026, that Meta’s strategy for enterprise AI goes far beyond its previously anticipated scope. According to TechCrunch AI’s Sarah Perez, in addition to the enterprise AI agents announced in June, Meta is exploring opportunities in API offerings, business agent sales, direct computing resource sales, and even external distribution of its internal tools for large-scale clients.

Moving Beyond Advertising Dependency

Zuckerberg addressed significant enterprise opportunities during the earnings call, emphasizing sales of APIs, business agents, compute resources, and services designed for larger clients. This shift could allow Meta to generate new revenue streams and reduce its dependence on advertising, which currently constitutes the majority of its income, alongside limited subscription revenue.

Initially, Meta plans to focus on its existing advertiser base by providing AI agents for messaging apps. Over years, Meta has built relationships with millions of advertisers and hundreds of millions of small businesses. Zuckerberg explained, “Just like our advertising system, we’ll earn revenue once we deliver results for these businesses,” positioning this as a natural extension of Meta’s established sales and partnership approach.

Expanding AI Agents

In June 2026, Meta introduced enterprise AI agents aimed at automating customer service, support, and daily operations. However, Zuckerberg’s comments during the earnings call reveal ambitions that extend well beyond these initial offerings.

The CEO also hinted at the possibility of making Meta’s internally-developed productivity tools and coding solutions available to external customers. “We’ve built coding, development, and internal productivity tools—partly because we needed them ourselves. Now that we’ve optimized these tools for internal use, we see significant opportunities to offer them to both small and large businesses.”

Compute Sales Strategy

Another focal point is the direct sale of compute resources. Meta currently possesses substantial GPU assets used for training and running AI models. Zuckerberg acknowledged, “There’s an opportunity to sell compute resources at a premium far exceeding the price we paid for them.”

However, Zuckerberg also emphasized caution in prioritizing short-term profits over long-term goals. “Selling off all compute resources for short-term gains would be foolish,” he noted, laying out plans to manage compute infrastructure as a “portfolio” combining short- and long-term objectives. He stressed the importance of seamless hardware, particularly for achieving “personal superintelligence,” and reaffirmed Meta’s commitment to investing in foundational models.

Challenges in Enterprise Sales

Zuckerberg admitted that enterprise sales would require Meta to develop entirely new expertise, describing it as “a different muscle than what we’ve built so far.” As a consumer-facing platform company, Meta’s ability to create customized sales processes and support frameworks for large enterprises will be critical moving forward.

Historically, Meta has provided self-service platforms for advertisers. Enterprise sales, however, demand capabilities such as long-term contract negotiations, dedicated customer success teams, and compliance with strict regulatory standards. Addressing these gaps will be a key determinant of the strategy’s success.

Major Investments in Agentic AI

The earnings call also highlighted Meta’s ambitious plans for agentic AI—systems capable of executing actions autonomously for individuals and organizations. Unlike traditional chatbots that merely respond to queries, agentic AI focuses on completing tasks independently.

Meta has already open-sourced its Llama series of large language models, using them as the foundation for enhancing agent capabilities. Enterprise AI agents will operate on platforms like Messenger and WhatsApp, automating customer interactions. In the future, Meta plans to offer agents capable of autonomously handling more complex business processes.

Industry Implications

Meta’s entry into enterprise AI is a move that competitors cannot ignore. Microsoft has bolstered its enterprise AI offerings with the Copilot series, Amazon Web Services (AWS) provides Amazon Q enterprise AI agents, and Google Cloud targets the same market with Vertex AI Agent Builder.

Meta’s advantage lies in its massive user base and well-established relationships with millions of advertisers. By integrating AI agents with its advertising platform for small businesses, Meta could lower adoption barriers. However, fulfilling the sophisticated requirements of large enterprises will likely demand additional investments in sales infrastructure and compliance capabilities.

In compute sales, Meta’s ability to balance its own AI model development with customer-driven resource allocation will be closely watched. Amid ongoing GPU shortages, Meta’s significant GPU holdings could play a pivotal role in shaping the industry’s resource distribution.

Editorial Opinion

In the short term, Meta’s model of offering performance-based AI agents to advertisers leverages its existing strengths and appears to be a practical approach. If hundreds of millions of small businesses begin using AI on Meta’s platforms, it could create new monetization channels alongside advertising revenues. However, Meta’s ability to quickly adapt to the demands of enterprise sales—an area requiring entirely different capabilities—remains uncertain. Competing firms with established enterprise sales channels may outpace Meta unless it strengthens its “different muscle” rapidly.

From a long-term perspective, compute sales and external distribution of Meta’s internal tools have the potential to fundamentally transform the company’s business structure. Selling compute resources, much like cloud providers, could become a high-margin, stable revenue stream. Yet, challenges remain in balancing trade-offs between customer sales and Meta’s internal AI research. As Zuckerberg’s mention of “personal superintelligence” suggests, consumer-facing next-gen AI experiences remain a priority for the company.

References

• “Zuckerberg says Meta’s enterprise AI opportunity extends beyond agents”, by Sarah Perez — TechCrunch AI, 2026-07-29T22:23:12.000Z (ARR)
• Source URL: https://techcrunch.com/2026/07/29/zuckerberg-says-metas-enterprise-ai-opportunity-extends-beyond-agents/