AI labs face severe talent bottlenecks: Anthropic retained nearly all staff during a Meta poaching wave while safety org Metr can staff only 2–3 of 20 priority projects.

Metr has 20 high-priority projects it wants to execute but capacity to staff only 2-3 of them, indicating a severe engineering talent bottleneck at a leading AI evaluation organization.

“They have 20 really useful projects they'd love to do but they only have capacity to do two or three of them.”

When Meta and other labs offered large compensation packages to poach technical talent, Anthropic lost only two employees while other labs lost dozens.

“When you know Meta and others were out you know with these huge packages for some of the technical talent in across the large language labs I think we lost two people and other labs lost dozens.”

Senior Cursor engineers have already quit and gone to X.AI.

“Senior engineers have already quit and gone to XI.”