Anthropic and the Danish drugmaker Novo-which recently changed its name from Novo Nordisk-have formed a partnership to deploy Claude AI in drug discovery. The collaboration targets specific bottlenecks in the R&D pipeline, aiming to shorten the timeline from early research to new treatments reaching patients.
According to Novo, the companies will jointly address key drug discovery challenges identified by scientists and computational teams at Novo. They plan to develop targeted solutions for specific scientific workflows, supporting biological reasoning and accelerating the development of new medicines. As an initial step, Novo will test Claude Science in select R&D workflows where the combined expertise of both organizations is expected to have the greatest impact.
What the partnership targets
Anthropic's President, Daniela Amodei, wrote on LinkedIn that Novo's experts have "pinpointed challenges in the drug discovery process that Claude is particularly capable of addressing." The work will focus on improving outcomes across the research process. Novo also said it will use Anthropic's frontier models to strengthen AI-driven software development internally.
Mike Doustdar, President and CEO of Novo, said, "AI can help us increase productivity in R&D and compress the path from research to marketed product." He added that AI tools can also open up new scientific opportunities and aid reasoning about human biology and drug mechanics.
Claude's recent scientific performance
The partnership follows findings Anthropic presented in August, which showed Claude's performance in a protein design campaign. The model designed protein binders against a range of targets, performing as well as-or, Anthropic claims, better than-leading human experts. The AI lab also demonstrated how Claude Opus 5 handled an analytical chemistry task, showing how general-access models can support routine, time-intensive research.
Dario Amodei, Co-Founder and CEO of Anthropic, framed the broader ambition: "AI's increasing capability brings with it the potential to compress a century's worth of biological and medical breakthroughs into a decade." He said giving leading researchers access to safe, capable frontier models can shorten timelines, improve outcomes, and discover medicines that significantly improve human life.
The broader AI and drug discovery context
This is not Novo's first AI partnership. In April, the firm announced a strategic collaboration with OpenAI to apply advanced AI to complex datasets, identify promising drug candidates, and reduce the time from research to patient. The latest deal with Anthropic deepens Novo's bet on frontier models across its R&D operations. For those tracking the use of AI for Science & Research, the partnership offers a concrete case study of how large models are being integrated into pharmaceutical workflows.
AI's role in drug development has shifted from a supportive analytical tool to a computational strategy used to assist therapeutic discovery. A recent article in Nature by Wonbeak Yoo outlined translational advantages that AI-enabled approaches have demonstrated: acceleration of early-stage timelines, reduction in candidate attrition, and enhancement of late-phase development. The field saw a landmark moment in November 2020 when Google DeepMind's AlphaFold was recognized as a solution to the 50-year protein-folding problem, predicting protein structures in minutes with remarkable accuracy.
That breakthrough earned a Nobel Prize for Demis Hassabis and John Jumper. Jumper joined Anthropic earlier this year, moving from Google DeepMind to its AI for Science team-a personnel shift that adds weight to Anthropic's scientific ambitions and the capabilities behind models like Claude AI Courses.
Why this matters for science and research professionals
The Novo-Anthropic partnership signals a practical shift: frontier models are moving beyond proof-of-concept studies and into active pharmaceutical R&D pipelines. For researchers and computational scientists, the key takeaway is that general-access AI models are now being stress-tested against domain-specific benchmarks-protein design, analytical chemistry-and performing at or above expert level. That changes the baseline for what in-house teams should expect from AI tools, and it raises the bar for evaluating which workflows can be meaningfully accelerated. Professionals who understand where these models fit into the drug discovery process will be better positioned to lead adoption rather than react to it.
Your membership also unlocks:
