Meta CEO Mark Zuckerberg: Everyone Will Have a Personal AI Assistant in Five Years
Meta CEO Mark Zuckerberg told investors during the quarterly earnings call that looking ahead five years, it is highly unlikely that billions of people will not have a personal assistant that understands their goals and acts on their behalf around the clock, covering finance, health, relationships, and family management. He later described the same product in a company blog as "extraordinarily capable, understanding you, your goals, and everything you care about," and stated that this is the foundation for the next wave of products and revenue streams.
The entry point is designated as existing messaging applications. He mentioned that people will interact with multiple assistants simultaneously, and that WhatsApp and other messaging interfaces will become more important, with WhatsApp already being the primary platform for users to access Meta AI. The assistant for businesses has been adopted by over 1 million companies after its global rollout on WhatsApp and Messenger. He marked intelligence as a business with higher margins than directly selling computing power, while computing power itself will continue to focus on large client businesses.
He used his own schedule as an example: the assistant monitors sleep, observes training and provides feedback, plans baking recipes for his eight-year-old daughter, orders ingredients, and gives advice in the kitchen, while they also design robots together; glasses are described as an interactive interface that keeps people present, with privacy compared to WhatsApp encryption. Muse is launched as a consumer-grade personal assistant. In the same call, he acknowledged that personal assistants have not yet reached the described scale, but believes it is almost inevitable that someone will succeed.
The bill appears concurrently. Meta's AI-related investments for the quarter were reported at around $31.1 billion, with free cash flow significantly declining, and after-hours stock prices fell by about 10%, while Reality Labs continues to incur substantial losses. Computing power is allocated to training models, expanding core business, delivering personal assistants, and selling computing to large clients. The Super Intelligence Lab and TBD group have written "personal superintelligence for everyone" as a different path from enterprise-level labs: if only labs serving companies and governments lead, the power will lean towards institutions rather than individuals.
Personal assistants aim to integrate sleep, training, children's extracurricular activities, and family systems into the same context. Sherry Turkle, founder of MIT's Technology and Self Initiative, described this as bringing AI into the most intimate moments, rather than just being a tool. Regulators and the public are still questioning whether the same company can be entrusted with much less sensitive data. Product promises encryption isolation, but the business model aims to turn understanding you into the next wave of revenue.
In market mechanisms, buyers are consumers needing around-the-clock assistance and businesses wanting messaging entry points, while the seller is Meta, which integrates assistants into WhatsApp, Messenger, and glasses. The event-driven aspect is that on earnings report day, a consumption scenario must be provided for the $100 billion computing expenditure. Funds flow from advertising profits and bonds into data centers, then back through assistant subscriptions or advertising increments. Beneficiaries are platforms controlling conversation entry points and home sensors; those under pressure are model companies that only do chat boxes, lack social graphs and devices, and must hand over sleep and parenting data to the platform to enjoy "understanding you."
Source: Public Information
ABAB AI Insight
Zuckerberg's shift from the unfulfilled "next platform" of the metaverse to personal assistants presents investors with a five-year inevitability ticket. The social graph already knows who you talk to, while glasses and wristbands need to know how you sleep, train, and raise children; assistants merge these two data sets into a single task authority. Using WhatsApp as an entry point equates to covering the deepest context collection with an encryption narrative. Selling intelligence has higher margins than selling computing power; this statement reflects a revenue model, not a philosophy.
The capital path involves the advertising machine providing blood transfusions to data centers. The contraction of free cash flow and stock price punishment indicates that the market has not yet accounted for "billions of personal assistants" in current profits. The 1 million business assistants represent a B-end landing, while the C-end's 24-hour understanding remains a demonstration and lengthy text. The Super Intelligence Lab is hiring high salaries, but the product side needs to make intimate life a default activation item. Whoever controls the messaging thread controls the assistant's work memo.
The analogy is not a ChatGPT dialogue box, but Facebook turning the social graph into an advertising operating system: first free, then around-the-clock, and then monetizing understanding. Apple differentiates through device privacy, Google through search and assistants, and Meta through chat and glasses. The industry position is in the land-grabbing phase of consumer assistants: model capabilities are leveling off, while entry points and sensors become scarce. Once intimate understanding becomes standard, the cost of switching assistants will exceed that of switching phones.
Structural judgment belongs to the overlay of technological substitution and pricing power transfer. What is being replaced is the part of life management that involves self-remembering, self-scheduling, and self-monitoring health; pricing power shifts from users' attention to the platforms that the continuously operating assistants depend on. The mechanism is that 24-hour assistance must be predicated on 24-hour context; once the context settles, the platform gains negotiation power for the next renewal and advertising. Understanding you means locking you in.