Meta had a plan called Project OT that would have cut team sizes by 60% through massive layoffs and AI automation. In January, Zuckerberg and his leadership team sketched out this vision at his Hawaii compound—smaller "talent-dense" teams of 3-5 people would handle work that previously required 10-20 engineers, with AI doing much of the heavy lifting. The company projected this would require layoffs bigger than the 25% cut they'd already done in 2022-2023. They planned two waves: one in May and another in November. But on May 19, hours before the first layoff, Zuckerberg backed off. He went through with a 10% cut the next day but canceled the November phase entirely. By then, Meta employees had figured out what was happening—the AI initiatives weren't about boosting productivity, they were about replacing them. The thinking behind Project OT came from what Meta executives saw in Asia. Zuckerberg's team visited startups there and noticed how they'd built lean organizations around AI. They commissioned research, ran pilots with "AI-native pods," and became convinced that this was the future. The math seemed straightforward: if a 5-person team with AI tools could do what a 15-person team used to do, why keep the extra bodies? The problem is that Meta wanted this transformation to happen in a year through brutal restructuring, not organically over time like actual startups experience. Some teams still got hit with 30-40% cuts and couldn't handle the workload. Worse, when experienced engineers got reassigned to AI data labeling, critical domain knowledge walked out the door. The real costs of going this small get glossed over in the "AI-native" excitement. A 3-5 person team has almost no redundancy—if someone's sick or on vacation, you're screwed. You can't maintain a proper on-call rotation with fewer than six engineers. There's no slack time for innovation, recruiting, or cross-team collaboration. Junior engineers don't get mentored as much. And you lose the diversity of judgment that comes from experienced people debating decisions. Meta's actual business was printing money with record revenue and profits, so the urgency doesn't make sense unless Zuckerberg is genuinely paranoid that some startup will out-execute Meta the way Facebook once destroyed Myspace.