Neolabs face 1-in-10,000 success odds under dual breakthrough-plus-scale mandate

The post identifies a structural mismatch: frontier labs prioritize scaling proven architectures while researchers seek paradigm shifts. Senior researchers therefore spin out neolabs at multibillion valuations with no product or revenue. These entities are simultaneously expected to produce AGI-level jumps and later-stage commercialization, splitting internal focus and eroding startup advantages in alignment.

Pharma data shows the division of labor works. From 2013-2022, small biotechs under $500M revenue originated a rising share of FDA-approved drugs; larger firms then acquire or license after proof-of-concept. The post cuts off mid-sentence but cites this pattern as the template for AI: neolabs perform the high-risk discovery phase and exit via acquisition rather than independent scaling.

Operationally this implies neolab term sheets should embed explicit acquihire or IP-sale triggers at defined capability thresholds instead of revenue milestones. Investors accept lower multiples in exchange for defined exit paths, while researchers avoid product and GTM responsibilities. The model decouples the two low-probability events without requiring neolabs to master both.