Meta Scraps Plan to Replace Staff With AI Agents
Meta shelved an internal plan to replace a large share of its workforce with AI systems after employee backlash and disappointing results from its AI agents, according to Reuters, as reported by the-decoder.com.
Internal documents cited by Reuters describe a program codenamed “Project OT” that aimed to shrink many teams by as much as 60 percent, leaving small groups of human staff to supervise virtual AI workers. CEO Mark Zuckerberg halted a planned second wave of cuts on the evening of May 19, just hours before the first round of layoffs was set to begin, and later called off a further round planned for November.
Reuters reports several factors drove the reversal. The AI agent technology did not deliver the productivity gains Meta had expected, investors criticized the company’s large AI spending, and employees pushed back openly. Staff came to believe that software tracking their mouse clicks and keystrokes was being used to train AI systems meant to replace them, prompting a wave of angry posts on Meta’s internal network, according to Reuters.
Internal sentiment measures at the company reportedly fell from 74 percent to 55 percent during the episode. In July, Zuckerberg acknowledged that the agent technology had not advanced as quickly as he had anticipated, the-decoder.com noted.
The episode highlights the gap between ambitions for AI-driven workforce automation and the current capabilities of AI agents, as well as the internal resistance such plans can generate among employees who fear being displaced by the same tools they are asked to help train.
Based on reporting by the-decoder.com.
