Meta tried to reorganize parts of its workforce around AI agents and substantially smaller teams, only to pull back after the technology failed to deliver the expected productivity gains, according to a Reuters investigation.
The effort, code-named Project OT for “Organization Transformation,” envisioned an ‘AI-native’ Meta in which autonomous agents performed much of the work now done by employees while smaller groups of workers oversaw them.
Executives considered scenarios that would shrink some teams by as much as 60%, Reuters reported, citing internal documents and people familiar with the project.
But internal data showed autonomous AI agents were not producing the productivity improvements executives had anticipated, according to Reuters.
Meta’s internal data showed AI-generated code volume rising dramatically without a comparable increase in useful output. Code changes to internal platforms and infrastructure were up 220% year over year, while changes resulting in new or improved features reaching users rose only 36%.
The initiative also encountered employee resistance, workflow disruptions and deteriorating morale.
Employee sentiment also fell from 74% favorable to 55% favorable in Meta’s half-year Pulse survey. Meanwhile, headcount in some engineering units dropped as much as 30% by the end of May, through a combination of layoffs and reassignments.
Meta ultimately carried out a roughly 10% workforce reduction in May but abandoned a planned second round of cuts slated for November.
More strikingly, internal posts said major technical and security incidents – including service disruptions and possible data leaks – increased 40%, while employees’ time spent “firefighting” those problems increased 70%. Infrastructure teams had reportedly warned about reliability problems associated with the surge in AI-generated code.
Meta confirmed the existence of Project OT and told Reuters it was a yearlong effort focused on cutting costs, redesigning teams and moving employees into priority areas. The company also acknowledged that its most aggressive scenarios contemplated reductions of as much as 60% in some teams.