Zuckerberg Admits Meta's AI Agent Rollout Is Behind Schedule — What It Means for Enterprise AI

Mark Zuckerberg told employees at an internal town hall that Meta's AI agents have not advanced as quickly as leadership expected — a rare public acknowledgment of a major AI strategy stumble from one of tech's most influential executives.

The admission follows Meta's decision in May to cut roughly 8,000 jobs (about 10% of its global workforce) and redirect 7,000 more employees to AI-focused teams. Zuckerberg now says the "trajectory of the agentic development over at least the last four months hasn't really accelerated in the way that we expected."

He told staff that executives had been "super optimistic" earlier this year about tools like Claude Code from Anthropic, and that fear of falling behind drove the January-February planning that led to the restructuring. That restructuring, he admitted, wasn't as "clean" as it should have been.

Key Takeaways:

• Zuckerberg expects meaningful benefits from Meta's AI investments within three to six months — setting a clear internal accountability window.
• Meta CTO Andrew Bosworth acknowledged the Applied AI division rollout was "atrocious," with the 6,500-person unit described internally as chaotic.
• Meta is projected to spend up to $145 billion on AI infrastructure this year despite the setbacks.
• Bosworth reversed course on employee monitoring software, shifting to opt-in after backlash.

The admission is a significant data point for AI strategy leaders: even the most well-funded organizations are discovering that agentic AI deployment timelines are harder to predict than anticipated.

Read the full article on The News International