Eli Lilly is paying $2.75 billion (including $115M upfront) to Insilico Medicine for exclusive global rights to develop and commercialize AI designed oral drug candidates, including at least one GLP 1 receptor agonist targeting the same diabetes market currently dominated by…
Eli Lilly is hedging against Eli Lilly.
That is the most direct way to read the $2.75 billion deal the Indianapolis drugmaker signed with Insilico Medicine on Sunday. Lilly makes tirzepatide — sold as Mounjaro for diabetes and Zepbound for obesity — one of the most commercially successful pharmaceutical products in history. Now it is paying $115 million upfront, with milestones that could take the total deal value to about $2.75 billion, to a Hong Kong-listed AI drug company for the right to develop and sell whatever oral drug candidates Insilico puts forward.
The Financial Times, citing unnamed sources, reported that at least one of those candidates is a GLP-1 drug for diabetes. That is the same therapeutic market Mounjaro already dominates. Read in that light, the deal is not Lilly outsourcing drug discovery to AI. It is Lilly buying optionality on the next generation of its own franchise — before a competitor finds it first.
The deal structure, confirmed in a Hong Kong Exchange filing: Lilly receives an exclusive global license to develop, manufacture, and commercialize preclinical oral drug candidates across disease areas Insilico has nominated. Insilico gets $115 million up front and is eligible for development, regulatory, and commercial milestone payments that could take the total deal value to about $2.75 billion, plus tiered royalties on future sales.
The financial terms are large. The strategic logic is legible. Mounjaro's weekly injection format leaves room for an oral alternative — easier to take, potentially cheaper to manufacture, accessible to different patient populations. If Insilico's AI-discovered GLP-1 candidate reaches market, Lilly distributes it. If it fails, Lilly is no worse off than before.
"The best drug ever invented by humans."
That is how Insilico CEO Alex Zhavoronkov described tirzepatide in a STAT News interview published alongside the deal announcement. He said he has been on Mounjaro for a year. He also said he wants to develop the next one. On Sunday, Lilly agreed to fund that ambition.
Zhavoronkov was careful not to name the licensed asset, but Insilico's pipeline webpage was recently updated to note that a GLP-1 candidate has been out-licensed to an undisclosed partner. The deal announcement confirmed the focus is oral therapeutics — a format where tirzepatide, semaglutide, and other GLP-1 agonists have made their clearest manufacturing and distribution gains.
The deal fits a broader pattern in pharma: large companies paying to access AI-generated pipelines rather than building their own. Insilico has done licensing deals with Sanofi and with Menarini Group/Stemline, among others. The company's pitch has consistently been that its generative AI platforms — Chemistry42 for small molecules, Inclusiv for biologics — can produce candidates faster and cheaper than traditional discovery. Lilly's signature on a $2.75 billion check suggests that pitch has a customer it cannot refuse.
For the biotech industry broadly, the deal is a data point in an ongoing argument about whether AI drug discovery can produce candidates that survive clinical development. Insilico has advanced several programs into human trials. None have reached market. The Lilly deal does not change that — but it puts real money behind the proposition that Insilico's candidates are worth betting on.
That bet is also a signal about the GLP-1 arms race. Novo Nordisk, Amgen, Pfizer, and a roster of smaller companies are all chasing the next generation of obesity and diabetes drugs. Oral formats, longer durations, combination approaches, new targets. Viking Therapeutics is running Phase 3 trials for its subcutaneous GLP-1/GIP dual agonist VK2735 — enrollment in the VANQUISH-1 trial completed in November 2025 — while separately developing an oral tablet formulation that completed Phase 2 testing in August 2025 with mixed results; Phase 3 readiness for the oral program remains to be established. Roche's partnership with Zealand Pharma is worth up to $5.3 billion, and Roche separately reported 22.5 percent weight loss for its own subcutaneous GLP-1/GIP candidate in January 2026. Lilly's own pipeline includes next-generation candidates, and the company has been transparent about the need to protect its franchise as exclusivity on tirzepatide eventually erodes. The Insilico deal is the most recent expression of that urgency.
The deal was announced Sunday. The HKEX filing is here. The Financial Times first reported the terms last week, citing sources.