Mark Zuckerberg has made a bold prediction: within the next five years, billions of people will rely on personal artificial intelligence agents that understand their goals and work on their behalf around the clock. The Meta founder and chief executive made the statement during the company's quarterly earnings call with investors on Wednesday, positioning AI agents as the next major computing shift and a central part of Meta's future product and revenue strategy.

“I think that it's extremely unlikely if you look out five years from now... that you don't have billions of people with a personal agent that understands your goals and that is just working on your behalf 24/7 to achieve your goals in whatever the domain is that you care about,” Zuckerberg said. He described a wide range of possible uses, including managing finances, monitoring health, improving interpersonal relationships, and handling household chores.

The comments came as Meta tries to convince Wall Street that its heavy spending on artificial intelligence will eventually pay off. The company has poured billions into AI infrastructure, model development, and consumer products. Zuckerberg's vision is not limited to a single product. He said he expects people to interact with multiple agents in the future, and he argued that Meta's messaging platforms, especially WhatsApp, are already becoming central to that experience. “As we move toward a future where we're all interacting with multiple agents, I think that WhatsApp and our other messaging surfaces are going to become increasingly important,” he said. According to Zuckerberg, WhatsApp is already the leading platform where users interact with Meta AI.

Key facts

• Mark Zuckerberg predicts billions of people will have personal AI agents within five years.
• The agents could help with finances, health, relationships, and household management.
• Meta's stock dropped almost 10% after the company reported quarterly earnings.
• Reality Labs lost about $4.6 billion in the quarter, bringing total losses to roughly $88 billion since 2021.
• Meta's free cash flow fell 91% year over year to $784 million.
• Meta and BlackRock announced a partnership to build a $14 billion data center in El Paso, Texas.
• More than one million businesses have adopted Meta's business agents on WhatsApp and Messenger.

The race to build agentic AI

Meta is not alone in setting high expectations for AI systems that can act on a person's behalf rather than just answer questions. Google has emphasized custom AI agents as a key new feature in its Search overhaul, although the changes sparked outcry from users who felt overwhelmed by the constant presence of AI results. At the same time, subscriptions to Anthropic's Claude have skyrocketed as engineers praise Claude Code, an agentic coding assistant that can complete complex programming tasks with limited human supervision.

The concept of an AI agent is a departure from traditional chatbots. Instead of waiting for prompts, an agent is designed to pursue a goal independently, making decisions and taking actions across multiple tools and platforms. In Zuckerberg's telling, a personal agent should feel less like a search box and more like a digital employee, one that learns the user's preferences and works in the background. That vision raises important questions about privacy, data security, accountability, and the environmental costs of running billions of AI models.

Meta's AI spending draws investor scrutiny

Zuckerberg's long-term optimism is colliding with short-term financial realities. Meta's stock dropped almost 10% after the company posted its latest quarterly earnings, a sign that investors are worried about rising costs and uncertain returns. Meta's Reality Labs division, which builds augmented reality glasses, virtual reality headsets, and related software, lost about $4.6 billion in the quarter. That is roughly in line with losses the division has posted each quarter since 2021, bringing its running total to around $88 billion.

AI spending is likely to climb even higher. The company reported free cash flow of $784 million in the quarter, down from $8.55 billion in the same quarter last year. That represents a 91% drop year over year, fueled largely by investments in AI infrastructure. This week, Meta and BlackRock announced a partnership to build a $14 billion data center in El Paso, Texas. Such projects are intended to support the massive computing power needed to train and run advanced AI systems, but they also raise concerns about energy consumption and climate impact.

The company's financial discipline is under intense scrutiny. Meta has a history of making big bets on speculative projects. The decision to rename the company Meta and pivot toward the metaverse in 2021 was initially met with skepticism, and the Reality Labs losses continue to weigh on the company's financial statements. Now Zuckerberg is asking investors to trust that a new wave of AI products will produce meaningful revenue. “We believe that there will continue to be a significantly higher margin on selling intelligence rather than selling compute directly, but we think that there's a big opportunity, obviously, to sell compute as well,” he said.

Building on WhatsApp and business agents

Meta has already begun introducing AI agents for businesses. The company's business agents, rolled out globally on WhatsApp and Messenger this quarter, have been adopted by more than one million businesses. These agents are designed to handle customer service, answer questions, and process orders, allowing companies to scale support without hiring more people. The early adoption suggests that businesses see practical value in AI agents, even if consumer use is still in its early stages.

The next challenge is to get individuals to adopt personal AI agents in large numbers. Zuckerberg believes the path to billions of users runs through Meta's existing messaging apps. WhatsApp has more than two billion users worldwide, making it one of the most popular communication platforms on the planet. By integrating AI agents directly into that environment, Meta can lower the barrier to entry and reach users who might not seek out a separate AI product.

Zuckerberg also pointed to the broader shift in computing interfaces. He has repeatedly said that Meta AI is one of the most widely used assistants in the world, and he sees messaging as the natural home for agentic services. In his view, users will eventually talk with AI agents the same way they talk to friends and colleagues, and those agents will be able to coordinate with one another to complete tasks. That could mean an agent scheduling meetings, another agent booking travel, and another agent managing household inventories, all operating in the background of a chat app.

The social implications are significant. If billions of people hand off financial decisions, health monitoring, and relationship advice to AI agents, questions of trust and reliability become crucial. What happens when an agent makes a harmful mistake? Who is responsible if a financial agent makes a bad investment? How will personal data be protected when an agent has access to sensitive information? These are issues that regulators, companies, and researchers will need to address as agentic AI becomes more common.

Competition and market pressure

Meta faces intense competition from other AI leaders. Google is integrating agents into its search engine and cloud services. Microsoft has invested heavily in OpenAI and is adding agentic capabilities to its Copilot assistant. Amazon is developing its own AI models and services. Anthropic has gained a strong following among software developers with Claude Code, which has been described as a transformative tool for coding productivity. OpenAI has also introduced tools that can perform tasks beyond conversation, such as browsing the web and taking actions in software applications.

Zuckerberg argues that Meta has a unique advantage: distribution. The company's apps are used by billions of people every day, and Meta AI is already embedded in Facebook, Instagram, WhatsApp, and Messenger. He says that gives Meta the ability to introduce AI agents at a scale that smaller competitors cannot easily match. Still, the company's heavy spending and mixed track record on moonshot projects make some investors cautious.

The phrase “selling intelligence” reflects a broader belief that the value in AI will not be in hardware or raw computing power, but in the models and services that can think, reason, and act on behalf of users. Zuckerberg wants to build the platform where that intelligence is delivered. He has also made Meta's large language models open-source in an effort to attract developers and create an ecosystem around Meta's AI technology.

The company's commitment to open-source AI has been controversial. Some researchers warn that powerful models could be misused if released freely, while others argue that open development is necessary to prevent a small number of corporations from controlling AI. For Meta, open-source models also serve a strategic purpose: they can become the standard that other companies build on, even if Meta is not the primary seller of every AI service.

Data centers and climate costs

The growth of AI has a physical side. Building the large language models that power personal agents requires enormous clusters of servers, specialized chips, and reliable electricity. Meta's plan to build a $14 billion data center in El Paso with BlackRock is a sign of how large the infrastructure investment has become. The facility is expected to support a range of AI services, but it also adds to the company's capital expenditure and operating costs.

Energy consumption is a growing concern for the entire industry. Data centers already account for a significant share of global electricity use, and AI workloads are far more intensive than traditional cloud computing. If Zuckerberg's prediction comes true and billions of people maintain personal AI agents, the environmental cost could be enormous unless there are breakthroughs in energy efficiency and renewable power. Zuckerberg acknowledged the need for better data center technology, but he did not provide a detailed plan for limiting climate damage.

Meta's financial results hint at why Zuckerberg remains focused on AI despite the short-term pain. The company's core advertising business still generates substantial cash, but growth is facing headwinds from privacy changes, platform competition, and economic uncertainty. AI agents could open new revenue streams beyond advertising, including subscriptions, transaction fees, and business services. Meta has not yet announced exactly how it plans to monetize consumer agents, but Zuckerberg believes they will be “the foundation for our next wave of products and revenue lines in the months and years ahead.”

So far, Meta's business agents have been adopted by more than one million businesses, which gives the company a base from which to expand. Consumer AI agents are a harder sell, though. People may be reluctant to trust an AI with their finances or personal relationships, and the technology still struggles with logic, common sense, and unpredictability. The road to “billions” has to start somewhere, and the company cannot get there on enterprise agents alone.

Source: TechCrunch News