Or that they were more likely to take risks. It’s not clear to me that what anthropic did is a mistake, they didn’t have the same level of capital and OpenAI is taking extreme risks with their compute commitment. It’s really not obvious that won’t backfire