How Insilico Medicine and Lilly Are Automating the Drug R&D Stack
AI / Machine Learning | 5 min read
Insilico Medicine (HKEX: 3696), a clinical-stage biotechnology company powered by generative AI and automation, has announced a landmark drug discovery collaboration with Eli Lilly and Company — one of the largest AI-driven drug development deals to date. Under the terms of the agreement, Insilico will receive an upfront payment of US$115 million, with potential total deal value of approximately US$2.75 billion in development, regulatory, and commercial milestones, plus tiered royalties on future sales. The agreement grants Lilly an exclusive worldwide licence for the development, manufacturing, and commercialisation of novel oral therapeutics from Insilico's preclinical pipeline across multiple therapeutic areas — while the two companies also collaborate on additional R&D programmes using Insilico's Pharma.AI platform combined with Lilly's clinical development expertise.
"From its inception, Insilico Medicine has been developing deep learning for end-to-end drug discovery. By deploying frontier AI technologies that scale from biomarkers to life models, world models of human and animal life, we can identify multi-purpose targets driving multiple diseases at the same time. Working with Lilly, we aim to deliver transformative therapies that treat diseases with high unmet need. This collaboration is a testament to the power of AI in tackling the most complex challenges in human health."
— Alex Zhavoronkov, PhD, Founder and CEO, Insilico Medicine
"Insilico's AI-enabled discovery capabilities represent a powerful complement to Lilly's deep expertise in clinical development across multiple therapeutic areas. This collaboration allows us to explore novel mechanisms and accelerate the identification of promising therapeutic candidates across multiple disease areas."
— Andrew Adams, Group Vice President of Molecule Discovery, Eli Lilly
Pharma.AI: Automating the Entire Drug Discovery Pipeline
At the centre of the collaboration is Pharma.AI — Insilico Medicine's end-to-end AI suite designed to automate the entire process of making a drug, from selecting the right biological target to predicting whether the drug will actually work in humans. The platform enables a software-defined pipeline: merging generative AI with traditional drug development to create an automated, data-first R&D model where every stage — target identification, molecule design, clinical success prediction — is accelerated by machine intelligence. Together, Insilico and Lilly will build out this unified tech stack, combining Pharma.AI's automated discovery capabilities with Lilly's proven clinical development infrastructure and deep disease-area knowledge.
Insilico's AI-Discovered Pipeline: 28 Drugs, Half at Clinical Stage
Insilico has developed at least 28 drug candidates using generative AI tools, with nearly half already at a clinical stage — a remarkably broad AI-derived pipeline for a company that has spent 12 years developing its technology. Therapeutic focus areas include fibrosis, oncology, immunology, pain, and obesity and metabolic disorders. Among its disclosed candidates are a pan-KRAS inhibitor in oncology, Nav1.8 targeting pain, and CDK4 in cancer — alongside a portfolio of novel, previously undisclosed targets. Although the specific assets licensed to Lilly were not disclosed due to contractual restrictions, Zhavoronkov confirmed that Insilico had updated its pipeline page to note a GLP-1 candidate as out-licensed to an undisclosed partner — consistent with Lilly's deep expertise and commercial success in metabolic disease (Mounjaro/Zepbound). Insilico will also join Lilly's Gateway Labs biotech development community as part of the partnership.
The Broader Significance: AI and Chemistry at the Same Table
The deal represents a significant milestone in the pharmaceutical industry's transition toward an automated, data-first R&D model. Traditional drug discovery is slow, expensive, and failure-prone — with the average new drug requiring over a decade and billions of dollars to reach patients. Insilico's AI-native approach can synthesise molecules more quickly than traditional methods, identify multi-purpose targets driving multiple diseases simultaneously, and predict clinical success probability before significant wet-lab investment. For Lilly — itself a leader in applying AI to drug development, with Zhavoronkov publicly noting the company is "better in AI than Insilico in some areas" — this collaboration represents access to complementary AI-driven discovery capabilities that sit outside its own development model. The partnership was enabled by a prior software licensing agreement between the two companies in 2023, providing a foundation of technical interoperability for this expanded collaboration.
Key Takeaways
- • Insilico Medicine (HKEX: 3696) and Eli Lilly have announced a global R&D collaboration — US$115 million upfront, up to US$2.75 billion in milestone payments, plus tiered royalties — granting Lilly an exclusive worldwide licence to develop and commercialise novel oral therapeutics from Insilico's preclinical AI-discovered pipeline.
- • Pharma.AI, Insilico's end-to-end AI suite, automates the full drug discovery process — from target identification and molecule design through to clinical success prediction — creating a software-defined R&D pipeline that merges generative AI with traditional drug development.
- • Insilico has developed 28 AI-discovered drug candidates — nearly half at clinical stage — spanning fibrosis, oncology, immunology, pain, and obesity/metabolic disorders, including a pan-KRAS inhibitor, Nav1.8, CDK4, and a GLP-1 candidate out-licensed as part of this deal.
- • The collaboration extends beyond the asset licence: Insilico and Lilly will co-develop multiple additional R&D programmes on targets selected by Lilly, combining Pharma.AI's automated discovery with Lilly's clinical infrastructure — and Insilico joins Lilly's Gateway Labs biotech community.
- • The deal signals a broader industry inflection: the pharmaceutical R&D stack is becoming software-defined, with AI accelerating target identification, molecule synthesis, and clinical outcome prediction — and the future of medicine is as much about code as chemistry.