Isomorphic Labs near $2.1B raise

AI drug discovery just got a very expensive vote of confidence. Isomorphic Labs — the Alphabet-owned company spun out of DeepMind — said on May 12 that it closed a $2.1 billion Series B to scale its drug design platform and push more programs toward the clinic. (isomorphiclabs.com) That matters because this corner of biotech has had a nagging credibility gap for years. The software looked impressive, but actual medicines were slower to show up. This round says investors still think the payoff could be enormous. ### What is Isomorphic Labs, exactly? Isomorphic sits in the AI-for-drugs world, but it is not just selling a single prediction model. The company is trying to build a full drug design engine — a stack of models and lab workflows that can help identify targets, generate molecules, and improve them into real candidates. (biospace.com) Its pitch is basically that biology can be modeled well enough for drug discovery to become faster, cheaper, and more systematic. ### What happened this week? The company announced a $2.1 billion Series B on May 12. Thrive Capital led the round again after backing Isomorphic’s first outside financing, and Alphabet plus GV returned as existing investors. New money came from MGX, Temasek, CapitalG, and the UK Sovereign AI Fund. (isomorphiclabs.com) Isomorphic says the capital will fund its AI drug design engine, expand globally, and move its own therapeutic pipeline forward. ### Why is that number such a big deal? Because it is huge even by biotech standards, and especially huge for a company that is still early on the clinical curve. Isomorphic raised $600 million in March 2025 in its first external round. (isomorphiclabs.com) This new financing is more than triple that. One biotech trade outlet called it the second-largest biotech round ever, which gives you the scale here. ### What does Isomorphic already have in hand? It already has large pharma partnerships, which is the practical proof that big drugmakers think the platform is useful. In January 2024, Isomorphic signed multi-target small-molecule collaborations with Eli Lilly and Novartis that it said could be worth nearly $3 billion in milestones, excluding royalties. (isomorphiclabs.com) Then in February 2025, Novartis expanded its collaboration by adding up to three more research programs. ### So is this about software fees or making drugs itself? Both — and that is the interesting part. Plenty of AI-biotech firms either license tools to pharma or try to become fully fledged drug developers. (isomorphiclabs.com) Isomorphic is trying to do both at once. Partner programs bring validation and cash. An internal pipeline offers much bigger upside if the science works. The catch is that running your own pipeline gets expensive fast, which helps explain the size of this raise. ### Has AI drug discovery actually produced clinical proof? There are signs of progress, but the picture is still mixed. A useful recent example is zasocitinib, a TYK2 inhibitor now owned by Takeda after it was acquired from Nimbus, a company co-founded by Schrödinger. (isomorphiclabs.com) Takeda reported positive Phase 3 psoriasis data in late 2025 and presented fuller results in March 2026. But that is not an Isomorphic drug, and it does not settle the bigger question of whether end-to-end AI platforms can repeatedly produce approved medicines on their own. ### What is still missing? The simplest answer is disclosed clinical assets from Isomorphic itself. The company has said it expects its first AI-designed drugs to enter clinical trials by the end of 2026, but for now the market is still betting ahead of that milestone. (isomorphiclabs.com) That does not make the story fake — it just means this is still a financing-round story, not yet a drug-approval story. ### Bottom line? This raise does not prove AI has solved drug discovery. But it does show where serious capital is leaning. Investors are no longer just funding chatbots and coding tools — they are funding the idea that biology itself can become a computable design problem. (takeda.com) Isomorphic now has the money to try to prove that in the hardest possible arena: real medicines, real trials, and eventually real patients. (isomorphiclabs.com) (techstartups.com)