Chai Discovery raises $400M at $3.8B valuation as AI antibody design reaches pharma

Chai Discovery, founded by Joshua Meier as co-founder and CEO, announced today that it has raised $400 million（約640億円） in Series C funding, bringing its valuation to $3.8 billion（約6100億円） and total funding to approximately $630 million（約1000億円）. The round was led by Index Ventures and included Kleiner Perkins, Sequoia Capital, and Dimension, alongside new investors Bain Capital Ventures, Battery Ventures, Baillie Gifford, BDT & MSD, Sapphire Ventures, and Avra Capital. Existing investors including Thrive Capital, OpenAI, Oak HC/FT, Menlo Ventures, and General Catalyst participated in the round.

Chai Discovery builds frontier AI models designed to predict and reprogram molecular interactions, a core capability for accelerating drug discovery. The company's latest model, Chai-3, represents what the company calls a "step change" over its previous generation, Chai-2. Chai-3 reportedly doubles success rates for molecular interaction targets, achieving about 35% to 40% hit rates, and targets enhanced bonding affinity and broader antibody design foundations. Meier stated: "Tomorrow's medicines should be designed with the precision, speed and scale of modern engineering, and this support helps us move faster towards that future. AI drug discovery has moved from promise to deployment."

The company's commercial traction has accelerated sharply. Chai announced a landmark licensing agreement with Pfizer Inc., granting access to Chai-3 as well as an AI model trained on Pfizer's proprietary data. It has also signed a customer agreement with Eli Lilly and Co. and formed a formal collaboration with Novartis AG. These partnerships position Chai's technology directly within the drug-discovery workflows of three of the world's largest pharmaceutical companies.

Chai's focus on antibody design addresses a fundamental challenge in drug discovery: there are roughly a quintillion possible customizations that can uniquely target foreign bodies. Antibodies function as custom locks designed to identify and neutralize attackers, but identifying which ones work has traditionally required screening millions of potential molecules and testing them one by one. Chai's AI system circumvents this by performing rapid simulations based on disease targets to design molecules that fit correctly, eliminating the need to guess or test random combinations. Instead, the system curates the vast set of possibilities to generate likely high-quality candidates.

Despite this progress, the broader AI drug-discovery industry faces a significant validation hurdle. According to an executive report from Excelra Knowledge Solutions Pvt. Ltd., no AI-discovered drug has been approved yet, despite $20 billion（約3.2兆円） being poured into generative AI drug discovery. OncoDaily reports that more than 173 AI-originated drug programs are now in clinical development, up from almost two dozen in 2023, with 15 to 20 expected to reach working trials in 2026. However, clinical data reveals a slowdown: although AI-driven discovery achieves Phase I pass rates of 80% to 90%, the rate drops to about 40% in Phase II, matching traditional methods. Industry observers suggest that the path forward requires better understanding of how AI-designed candidates behave in clinical trials as they advance beyond initial safety testing.

Chai Discovery's $400 million（約640億円） Series C represents a major vote of confidence in AI-driven drug discovery as it transitions from laboratory proof-of-concept to commercial partnerships with tier-one pharmaceutical companies. The three-fold increase in valuation—to $3.8 billion（約6100億円）—reflects investor belief that the company's molecular-interaction prediction models can materially speed up one of pharma's costliest and slowest processes. The timing is significant: the company's CEO stated that "AI drug discovery has moved from promise to deployment," and the Pfizer, Eli Lilly, and Novartis partnerships underscore that belief by placing Chai's technology directly into production workflows at companies that collectively control billions in drug-development budgets.

However, the article also surfaces a hard truth: despite $20 billion（約3.2兆円） invested in generative AI drug discovery, no AI-designed drug has been approved yet. More than 173 AI-originated programs are in clinical development—up sharply from about two dozen in 2023—suggesting the pipeline is building. Yet the clinical data shows why this remains challenging: while AI achieves 80–90% pass rates in Phase I trials, the rate drops to 40% in Phase II, matching traditional drug discovery. This stall point suggests that AI's advantage in molecular design does not automatically translate to better clinical outcomes, and that Chai and its peers must help the industry gain deeper understanding of how designed candidates behave in human trials.