Zuckerberg: Billions Will Run Their Lives on AI Agents by 2031
Meta's CEO says personal AI agents working 24/7 are inevitable within five years — and he's spending up to $145B this year on it.
Mark Zuckerberg used Meta's Q2 2026 earnings call to make his boldest agent prediction yet: it's "extremely unlikely" that five years from now you don't have billions of people running a personal AI agent that knows their goals and works on their behalf 24/7 — across relationships, household, finances and health.
The delivery vehicle is messaging, not a new app. Zuckerberg says WhatsApp is already the top surface for Meta AI interactions and will be the front door for personal agents. He framed agents as the next big AI product category after coding assistants, and pitched "personal superintelligence" — echoed in a WSJ op-ed — as the anti-concentration alternative to a handful of labs owning the tech.
The number behind the talk: Meta raised 2026 capex guidance to $130–145B. That's the bet — that agents become infrastructure people depend on rather than a chatbot they open occasionally.
Why it matters: if Meta is right, the AI interface layer stops being a chat window and becomes a persistent agent with access to your calendar, money and relationships — and Meta wants to own it.
Sources
- Mark Zuckerberg predicts that billions of people will have personal AI agents in five years techcrunch.com
- Zuckerberg details Meta's personal AI superintelligence strategy artificialintelligence-news.com
Written by an AI pipeline from the sources above. How it works.
Feed, daily deep-dive and bytes — readable offline, with push alerts for the topics you follow.