
Leopold Aschenbrenner’s AI fund lost most of its value. Now it’s back with $400M.
THE SO WHAT
A fund that “nearly blew up” in July and is already redeploying $400M shows how much LP appetite remains for concentrated AI risk, even after visible drawdowns. If you’re an operator, expect more volatile, thesis-driven capital around you — funding cycles may compress, but expectations for step-change outcomes will spike.
READ THE SOURCE
MORE FROM THE WIRE
Startups & VentureInevitable AI Group closes $6M pre-seed round led by Aleph to bring SaaS to the AI age
A $6 million pre-seed for an "AI-native" SaaS venture studio reflects investor belief that many current SaaS categories will be rebuilt around agents and automation-first workflows. If you’re a SaaS operator, assume new entrants will design around AI from day zero and audit where your own product still treats it as an add-on.
Startups & VentureEx-Spotify employees raise $10M to bring the AI behind its recommendations to e-commerce
A $10 million round to port Spotify-grade recommendation engines into e-commerce is another step toward "taste graphs" for shopping. Retailers who don’t own their own real-time personalization stack will find margins squeezed as these systems become table stakes.
Startups & VentureDefense Startup Hadrian Valued at $7.87 Billion in New Round
A $7.87B valuation for Hadrian to scale defense manufacturing is more evidence that capital is re-rating hard-tech supply chains, not just software. If you sell into defense or aerospace, expect procurement to favor vendors who can plug into these new, software-driven factories on data and integration terms, not just price.
Startups & VentureSequoia Capital is putting $10bn behind AI and ‘reindustrialization’
Sequoia earmarking $10B for AI and “reindustrialization” is a capital allocation bet that software plus hard assets — factories, energy, logistics — is the next compounding frontier. Founders in anything that touches physical production should expect more receptive capital, but also higher expectations on defensibility and integration with existing industrial stacks.