Scientific discovery is the next bottleneck for intelligence
Post-training climbed maths, code and tool use on process data written by humans or synthesised under their direction. The research rung has no such corpus. To keep improving, models have to do science — which makes discovery AI's own bottleneck, not an application of it.