Mirendil Raises $200M to Speed Up Scientific Research With AI

My take: Read it twice because it left me thinking. Behnam Neyshabur and Harsh Mehta, former Google and Anthropic researchers, launched Mirendil with $200 million in seed funding (a startup's first big investment) at a $1 billion valuation, with no product or revenue yet. The idea is to build AI that helps scientists do better AI: accelerate the research, not replace the researcher. I love that, because it is exactly my thesis: AI came to empower whoever uses it, not to push them out of the game. The round was led by a16z, Kleiner Perkins, and Nvidia, real heavyweights. What catches my eye is that they left Anthropic, a house famous for its caution, to build something at this scale. I read it two ways: either they have something very concrete that nobody else has seen, or the seed-stage bubble still is not over 😅.