Business

Bristol Myers Squibb taps Chai Discovery's generative AI for antibody design across portfolio

Bristol Myers Squibb taps Chai Discovery's generative AI for antibody design across portfolio

San Francisco-based Chai Discovery and Bristol Myers Squibb (NYSE: BMY) have announced a collaboration to apply Chai's generative AI antibody discovery platform across BMS's broader portfolio, with the stated goal of building an AI-powered, continuously learning discovery system at BMS. The deal covers antibody candidate identification across multiple undisclosed targets. Financial terms were not disclosed.

Under the agreement, BMS gains access to Chai's molecular folding and design models, which predict and reprogram molecular interactions to generate novel biomolecule designs from defined therapeutic criteria. The platform's current-generation system, Chai-3, is a multimodal diffusion-based architecture capable of designing complete antibody sequences and structures from scratch conditioned on a target protein and binding epitope, with the company reporting that its predecessor Chai-2 achieved experimental hit rates of approximately 16–20% in fully de novo antibody design compared with sub-1% rates for prior computational methods.

No upfront payment, milestones, royalties, territorial rights, or equity component were disclosed.

The AllSci BriefSystematic R&D and deal news. Daily.

The BMS deal is the fifth major pharma platform agreement Chai has disclosed in under eight months, following collaborations with Eli Lilly (January 2026), Pfizer (June 2026), Novartis (July 2026), and argenx (July 2026). The BMS collaboration arrives roughly five weeks after Chai closed a USD 400 million Series C at a USD 3.8 billion valuation. BMS has separately been expanding its AI discovery infrastructure through a deployment of Schrödinger's (Nasdaq: SDGR) agentic platform Bunsen in August 2026 and a May 2026 enterprise agreement with Anthropic for Claude across its global operations.


Access the AllSci platform to explore the science behind the news.


Spot something wrong? Report an issue with this article