What is the story about?
Meta CEO Mark Zuckerberg has laid out one of the company's boldest predictions yet for artificial intelligence, arguing that personal AI agents will become a routine part of everyday life for billions of people within the next five years.
Speaking during Meta's latest quarterly earnings call, Zuckerberg said he expects AI to evolve beyond today's chatbot experience into software that actively understands an individual's priorities and continuously works towards helping them achieve personal goals.
"I think that it's extremely unlikely if you look out five years from now, for example — whatever period of time you want — 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 told investors.
According to him, these AI agents could eventually assist users with a wide range of everyday tasks, including managing finances, monitoring health, organising household responsibilities and even offering support in personal relationships.
The comments reinforce Meta's long-term strategy of embedding AI deeply into its products rather than treating it as a standalone tool. Zuckerberg also suggested that messaging platforms such as WhatsApp will play an increasingly central role as users begin interacting with multiple AI assistants throughout the day.
He noted that WhatsApp has already become Meta's leading platform for conversations with Meta AI, positioning the messaging service as a key gateway for future AI experiences.
The company's latest vision reflects a broader race across the technology industry, where leading firms are shifting their attention from generative AI that simply answers prompts to "agentic" AI capable of carrying out tasks on behalf of users.
Google has similarly introduced personalised AI agents as part of its overhaul of Search, although the rollout has faced criticism from some users who argue that AI-generated results have become intrusive. Anthropic, meanwhile, has seen strong demand for its Claude subscription services, particularly among software developers using its AI-powered coding assistant.
While Meta is positioning AI as its next major growth engine, investors remain focused on the enormous financial commitment required to build that future.
The company continues to pour billions into AI infrastructure, alongside investments in computing capacity and new hardware. This week, Meta and investment giant BlackRock announced plans to jointly develop a $14 billion data centre in El Paso, Texas, underscoring the scale of infrastructure needed to support increasingly sophisticated AI systems.
The spending, however, has begun to affect Meta's financial metrics.
The company reported free cash flow of $784 million for the quarter, a sharp decline from $8.55 billion during the same period a year earlier. The drop reflects the growing cost of expanding AI capabilities and building the computing infrastructure required to train and deploy future models.
Reality Labs, Meta's division responsible for augmented reality glasses, virtual reality headsets and related technologies, also remained deeply loss-making. The unit recorded another quarterly loss of roughly $4.6 billion, continuing a pattern that has persisted since 2021 and pushing cumulative losses to around $88 billion.
Despite those figures, Zuckerberg maintained that Meta sees significant commercial potential in AI. He argued that offering intelligent AI services would ultimately be more profitable than simply renting out computing power, although he acknowledged that selling compute itself could also become a meaningful business opportunity.
"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.
Meta has already begun commercialising AI through business-focused tools. The company said its AI-powered business agents, now available globally on WhatsApp and Messenger, have been adopted by more than one million businesses.
The bigger challenge lies in convincing everyday consumers to rely on AI agents as personal digital assistants. While enterprise adoption offers an early source of momentum, Meta's broader ambition depends on persuading billions of individuals that AI can become a trusted companion integrated into their daily lives. Whether that vision materialises may ultimately determine whether the company's massive investment in artificial intelligence delivers the returns Zuckerberg is betting on.
Speaking during Meta's latest quarterly earnings call, Zuckerberg said he expects AI to evolve beyond today's chatbot experience into software that actively understands an individual's priorities and continuously works towards helping them achieve personal goals.
"I think that it's extremely unlikely if you look out five years from now, for example — whatever period of time you want — 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 told investors.
According to him, these AI agents could eventually assist users with a wide range of everyday tasks, including managing finances, monitoring health, organising household responsibilities and even offering support in personal relationships.
Mark Zuckerberg sees AI becoming a personal assistant for every user
The comments reinforce Meta's long-term strategy of embedding AI deeply into its products rather than treating it as a standalone tool. Zuckerberg also suggested that messaging platforms such as WhatsApp will play an increasingly central role as users begin interacting with multiple AI assistants throughout the day.
He noted that WhatsApp has already become Meta's leading platform for conversations with Meta AI, positioning the messaging service as a key gateway for future AI experiences.
The company's latest vision reflects a broader race across the technology industry, where leading firms are shifting their attention from generative AI that simply answers prompts to "agentic" AI capable of carrying out tasks on behalf of users.
Google has similarly introduced personalised AI agents as part of its overhaul of Search, although the rollout has faced criticism from some users who argue that AI-generated results have become intrusive. Anthropic, meanwhile, has seen strong demand for its Claude subscription services, particularly among software developers using its AI-powered coding assistant.
Massive AI investments raise questions over Meta's spending
While Meta is positioning AI as its next major growth engine, investors remain focused on the enormous financial commitment required to build that future.
The company continues to pour billions into AI infrastructure, alongside investments in computing capacity and new hardware. This week, Meta and investment giant BlackRock announced plans to jointly develop a $14 billion data centre in El Paso, Texas, underscoring the scale of infrastructure needed to support increasingly sophisticated AI systems.
The spending, however, has begun to affect Meta's financial metrics.
The company reported free cash flow of $784 million for the quarter, a sharp decline from $8.55 billion during the same period a year earlier. The drop reflects the growing cost of expanding AI capabilities and building the computing infrastructure required to train and deploy future models.
Reality Labs, Meta's division responsible for augmented reality glasses, virtual reality headsets and related technologies, also remained deeply loss-making. The unit recorded another quarterly loss of roughly $4.6 billion, continuing a pattern that has persisted since 2021 and pushing cumulative losses to around $88 billion.
Despite those figures, Zuckerberg maintained that Meta sees significant commercial potential in AI. He argued that offering intelligent AI services would ultimately be more profitable than simply renting out computing power, although he acknowledged that selling compute itself could also become a meaningful business opportunity.
"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.
Meta has already begun commercialising AI through business-focused tools. The company said its AI-powered business agents, now available globally on WhatsApp and Messenger, have been adopted by more than one million businesses.
The bigger challenge lies in convincing everyday consumers to rely on AI agents as personal digital assistants. While enterprise adoption offers an early source of momentum, Meta's broader ambition depends on persuading billions of individuals that AI can become a trusted companion integrated into their daily lives. Whether that vision materialises may ultimately determine whether the company's massive investment in artificial intelligence delivers the returns Zuckerberg is betting on.
