Eli Lilly Taps Chai Discovery’s Frontier AI to Design ‘Computer-Aided’ Biologics from HIT Fred Pennic

OpenAI-Backed Chai Discovery Raises $130M to Tackle "Undruggable" Targets with Generative AI

What You Should Know

Eli Lilly and Company has entered a strategic collaboration with Chai Discovery to deploy frontier AI for the design of novel biologic therapeutics.

– The partnership includes the development of an exclusive, purpose-built AI model trained on Lilly’s proprietary data, leveraging Chai’s Chai-2 platform to compress discovery timelines from months to weeks.


The “Zero-Shot” Revolution: Slashing Discovery Timelines

In the world of biologics, antibody design has historically been a game of trial and error. Chai Discovery is disrupting this with Chai-2, the first zero-shot antibody design platform to achieve double-digit experimental hit rates.

By predicting and reprogramming biochemical interactions with “computer-aided design” (CAD) precision, Chai allows Lilly to:

  • Accelerate Hit Rates: Achieve high experimental success without the traditional “wet lab” cycles.
  • Design Drug-Like Properties: Ensure molecules are not just theoretically viable but possess the stability and safety required for human therapeutics.
  • Compress Timelines: Complete discovery processes in weeks that previously required months of manual experimentation.

Proprietary Data Training

The most significant component of this deal is the development of a custom AI model. While Chai offers access to its core frontier models, this purpose-built engine will be trained on Lilly’s large-scale proprietary data.

Skeptics often point to “AI Hype” in drug discovery. However, Chai Discovery’s recent $1.3 billion valuation (backed by OpenAI, Thrive Capital, and Oak HC/FT) suggests that the financial markets view their “hit rates” as clinically significant. Unlike point solutions, Chai is building a “frontier foundation model”—the biological equivalent of GPT-4—specifically for molecular interactions.

 Read More