Meta CEO Mark Zuckerberg’s ambitious plan to reshape the company around artificial intelligence has run into a major setback, after employee resistance and problems with AI technology forced the company to scale back a sweeping workforce transformation.
The initiative, internally known as Project OT, short for Organization Transformation, was developed earlier this year as part of Zuckerberg’s push to make Meta an “AI-native” company. The idea was to redesign how teams worked, automate large portions of daily tasks and rely on smaller groups of employees working alongside AI agents.
According to documents reviewed by Reuters and interviews with more than 20 people familiar with Meta’s operations, executives considered scenarios in which some teams could be reduced by as much as 60%. The plan was not necessarily a 60% cut across Meta as a whole; the company says the scenarios involved a combination of layoffs, redeployments and eliminating open positions.
The restructuring was initially expected to happen in two waves. The first took place on May 20, when Meta cut about 10% of its workforce. Thousands of other employees were moved into newly created or expanded priority areas, including AI-related work.
But the second phase never happened.
Just hours before the May layoffs, Zuckerberg called off planning for a second restructuring wave that had been expected in November. Reuters could not determine exactly what prompted the decision.
One major factor appears to have been growing employee resistance.
Workers had become increasingly concerned that Meta’s AI transformation was not simply about improving productivity, but about replacing their jobs. Internal Workplace posts reportedly included angry reactions and memes criticizing the layoffs, creating a serious morale problem inside the company.
At the same time, Meta’s AI systems were showing signs that they were not yet ready to replace large amounts of human work.
Internal posts indicated that AI-assisted coding dramatically increased the amount of code being produced, but the increase did not translate into an equivalent improvement in products reaching users. One internal post cited by Reuters said code changes to Meta’s internal platforms and infrastructure were up 220% year over year, while changes resulting in new or improved user-facing features were up only 36%.
Infrastructure teams also warned about reliability problems. AI agents were reportedly capable of taking unpredictable and disruptive actions, increasing the burden on employees who had to fix problems created by the systems. Internal posts indicated that major technical and security incidents, including service disruptions and possible data leaks, had risen substantially.
The backlash also forced Meta to reconsider some workplace practices associated with its AI push. The company paused a mouse-tracking program, allowed some employees who had moved into new AI engineering teams to return to their previous groups, and took steps to improve morale after the layoffs.
Zuckerberg subsequently told remaining employees that he did not expect further company-wide layoffs that year and said he wanted to provide greater stability.
Importantly, Project OT itself has not completely disappeared. Meta has described it as a broader, year-long effort focused on reducing costs, redesigning teams and moving employees toward priority areas such as AI development and training data. The company says it never intended to implement every scenario considered during the planning process.
The episode highlights a growing challenge for Silicon Valley: AI may be powerful enough to change how companies work, but it is not necessarily reliable enough to replace humans at the scale executives initially imagine.
For Zuckerberg, the lesson from Project OT may be that building an AI-native company requires more than cutting headcount and deploying AI agents. It also requires technology that can consistently deliver productivity gains without creating new technical problems, damaging employee morale or undermining the very teams needed to build and supervise the AI systems.
Meta continues to invest heavily in artificial intelligence. But after Project OT, the company appears to be taking a more cautious approach to the idea that AI can simply replace large numbers of human workers.