I think another third option is that perhaps there aren't many people with the skills to run large teams for both research and product. I suspect that DeepMind / OpenAI / Anthropic require infrastructure and tooling that's significantly more sophisticated than what smaller independent research/product teams have experience with.
replySo in addition to large-team leadership skills, there may be some niche technology/systems that only a limited number people have experience with multiple 1-3 year cycles of designing, building, maintaining, and replacing these uniquely large infra systems.
This is interesting but I'm not sure I get it. I can't tell if you're saying these folks leaving Google are part of that small group (I think you're saying they are, but why would them having those skills lead to them leaving? Because there are already other people at Google doing those roles and so they aren't making use of those unique skills?). I think I'm just not quite getting it, but I'd like to understand.
replyOne other thought from your comment, which has been beaten to death, is the legacy product problem for Google, which is not going to magically go away. For people on the bleeding edge they just don't want to have another meeting where someone brings up, for instance, search. You don't have to worry about legacy products at Anthropic and OpenAI because none exist so it's certainly more greenfield.