Replacing humans with AI doesn’t appear to be as straightforward as one might think, as evidenced by Meta’s situation.

According to Reuters, during an internal town hall on Thursday, CEO Mark Zuckerberg informed employees that the speed of AI agent advancement had not “picked up in the manner” that executives had previously anticipated.

Earlier this year, Meta terminated around 8,000 staff — roughly 10% of its overall workforce — and reassigned an additional 7,000 to various AI divisions, including one called Agent Transformation, Bloomberg stated.

In this week’s gathering, Zuckerberg reportedly remarked on these layoffs — indicating that they were not as “neat” as they ought to have been. The decisions were made because senior leaders at the company “were concerned that we weren’t moving quickly enough ‌to adjust” to the evolving dynamics of the tech sector, Zuckerberg allegedly stated.

The corporate head also reportedly mentioned that the anticipated benefits of the new AI-centric organizational structure had not “materialized yet,” though he expressed his belief that the company would start to notice advancements from its AI expenditures within the next three to six months. Several additional investigative accounts have illustrated Meta’s relatively new AI division as a demoralizing environment, according to some engineers involved.

Meta has made significant investments in AI and is projected to allocate up to $145 billion on AI infrastructure this year, Reuters reports.

TechCrunch reached out to Meta for a response.