Meta CEO Mark Zuckerberg is reportedly building an AI agent designed specifically to help him do his own job—and it’s already operational, albeit in limited form. That makes him the first big tech chief to move the conversation from philosophical to practical, after Google CEO Sundar Pichai told the BBC late last year that running a company might actually be one of the easier things for AI to eventually take over.
Zuckerberg ‘s ‘CEO agent’ is cutting through corporate layers
According to the Wall Street Journal, Zuckerberg’s so-called CEO agent is currently helping him retrieve information faster—answers he’d typically have to chase through multiple layers of staff to get. The tool is still in development, but it reflects a broader push at Meta to flatten its 78,000-person organisation and reduce internal bureaucracy through AI. Employees across the company have also been building their own personal agent tools—one called My Claw can access chat logs and work files, and can even ping colleagues on an employee’s behalf. Another, called Second Brain, built on top of Claude by a Meta staffer, functions like “an AI chief of staff,” per an internal post announcing it.
Big Tech CEOs are lining up to say AI could replace them
Pichai’s BBC comments from November 2025 kicked off the conversation. “I think what a CEO does is maybe one of the easier things maybe for an AI to do one day,” he said, adding that agentic AI—models that can act autonomously on a user’s behalf—would become far more capable in the next 12 months. He also acknowledged the flip side: job displacement is coming, and “people will need to adapt.” OpenAI’s Sam Altman had said something strikingly similar, telling an Axel Springer event that he’d be “nothing but enthusiastic” when AI becomes a better CEO of OpenAI than him. Even Klarna’s Sebastian Siemiatkowski posted on X that AI is “capable of doing all our jobs, my own included.” That said, Nvidia’s Jensen Huang pushed back hard—calling the idea that AI could take his job an absolute non-starter, and arguing that AI replacing workers at massive scale is still a long way off.What makes the Zuckerberg angle interesting is that it moves the conversation from philosophical to practical. Meta has formally tied AI tool adoption to employee performance reviews, and the internal atmosphere, per people familiar with the matter, feels reminiscent of the company’s chaotic, fast-moving Facebook-era culture. Some employees find that energising. Others are quietly anxious about what comes next.Whether AI can actually run a trillion-dollar company is still an open question. But at Meta, at least, the experiment has already begun.