Meta's plan to replace employees with AI falls apart

According to Reuters, Meta had been running an internal initiative codenamed "Project OT" (Organization Transformation), aimed at turning the company into what executives called an "AI-native" organization. The plan envisioned AI agents handling much of the day-to-day work currently done by employees, with much smaller human teams overseeing them.

Mark Zuckerberg reportedly discussed the project with top Meta executives earlier this year during a meeting at his property in Hawaii. Some scenarios under consideration involved cutting certain teams by as much as 60%, though Meta has stressed this referred only to specific teams under various restructuring scenarios, not the whole company.

Two phases, one halted

The changes were planned to roll out in two phases, in May and November. In May, Meta did lay off around 10% of its workforce. However, just hours before that round of layoffs, Zuckerberg decided to halt preparations for the second phase planned for November.

The decision came amid growing internal discontent, with employees feeling they were effectively being asked to help train technology that could eventually replace their own jobs.

AI results fell short of expectations

Internal data showed that AI significantly increased the volume of code produced, but this did not translate into a proportional number of real product improvements. At the same time, the number of technical and security incidents rose, forcing employees to spend more time fixing problems.

Zuckerberg later acknowledged that the development of AI agents had not progressed as quickly as he had hoped. This does not mean Meta is abandoning artificial intelligence — the company continues to invest heavily in it. Still, the Project OT experience has shown that replacing human labor with AI is proving far more complicated in practice than ambitious management plans suggested.