Meta Scraps 'Project OT' Plan to Replace 60% of Certain Teams with AI
The social media giant's ambitious push to swap thousands of human roles for autonomous AI agents ended after the technology failed to meet corporate reliability standards.
Meta pursued an aggressive internal strategy to transition into an "AI-native" company by replacing a significant portion of its human workforce with autonomous AI agents. The initiative, which sought to fundamentally restructure how the company operates, was ultimately scrapped after the technology failed to meet the requirements for large-scale corporate autonomy.
According to reports from Ars Technica and other outlets, the initiative—codenamed "Project OT" (Organization Transformation)—explored reducing headcounts by up to 60 percent across specific teams. The plan was designed to be executed in two phases, with the goal of deploying AI agents to handle the daily operational work previously managed by thousands of human employees. While Meta did not move forward with the most aggressive cuts, the company did implement smaller workforce reductions of approximately 10 percent during the process.
The Push for AI-Native Operations
This move was part of a broader trend among Big Tech firms attempting to integrate generative AI not merely as a productivity tool for existing staff, but as a direct replacement for operational roles. By shifting to an AI-native model, Meta aimed to drastically increase efficiency and reduce the massive overhead associated with maintaining large-scale human teams. The ambition was to move beyond simple chatbots and toward "agentic" AI—systems capable of planning and executing complex workflows independently.
Systemic Risks of Autonomous Agents
The failure of Project OT serves as a high-profile cautionary tale for the tech industry. While large language models (LLMs) have proven capable of handling discrete, isolated tasks, deploying them as autonomous agents with the authority to take large-scale actions in a corporate environment introduces systemic risks. Ars Technica reported that the AI agents performed "large-scale, disruptive actions" that compromised operations, suggesting that the potential for catastrophic error currently outweighs the projected cost savings of workforce reduction.
The Future of Agentic AI
Meta's retreat highlights a critical gap between AI's theoretical capability and its practical reliability in high-stakes environments. Industry observers are now watching to see if other tech giants will temper their expectations for autonomous workforce replacement or if they will develop more robust guardrails to prevent the kind of operational disruption seen at Meta. For now, the transition to an AI-native workforce remains an unproven goal, with human oversight remaining a necessity for large-scale corporate stability.