Meta’s AI agent progress slower than expected, Zuckerberg tells employees after major restructuring

Meta CEO Mark Zuckerberg has acknowledged that the company's efforts to develop AI agents are progressing more slowly than anticipated, even after a major restructuring aimed at accelerating its artificial intelligence ambitions. According to a Reuters report, Zuckerberg made the remarks during an internal town hall meeting on July 2.During the session, Zuckerberg admitted that Meta's sweeping reorganisation, which included significant job cuts and the reassignment of thousands of employees to AI-focused teams, had not gone as smoothly as expected. He said the company had misjudged the timing of some of the changes and acknowledged shortcomings in the execution of the restructuring.According to Reuters, Zuckerberg told employees that the "trajectory" of Meta's AI agent development over the last four months had not accelerated in the manner company leaders had anticipated. He added that some of Meta's key bets around its new organisational structure had yet to produce the desired outcomes.The comments come after Meta laid off roughly 10% of its global workforce, affecting around 8,000 employees, and reassigned approximately 7,000 workers to AI-related roles in May. The moves formed part of a broader plan to redirect resources toward artificial intelligence infrastructure and improve operational efficiency through AI-powered tools.Zuckerberg said company executives had entered 2026 with an optimistic view of how quickly AI agents would mature. According to the report, many senior leaders were concerned Meta might not move fast enough to keep up with the rapidly evolving AI landscape, particularly as tools such as Anthropic's Claude Code gained traction among developers.AI agents are designed to perform tasks autonomously on behalf of users and are viewed by many technology companies as one of the next major frontiers in artificial intelligence. Meta has made them a key component of its long-term AI strategy.Despite the slower-than-expected progress, Zuckerberg remained optimistic about the company's AI investments. He told employees that Meta is likely to see more meaningful returns from its AI initiatives within the next three to six months.Meta is projected to spend as much as $145 billion on AI infrastructure this year, making it one of the largest investors in the technology industry. The company's investment is part of a broader spending wave among major technology firms, which are collectively expected to invest more than $700 billion in AI-related infrastructure and development.The admission marks a rare public acknowledgement from Zuckerberg that the path to AI-driven productivity gains may take longer than initially anticipated, even as Meta continues to expand its AI capabilities aggressively and compete with rivals such as OpenAI, Anthropic and Google.