Novo has added Anthropic's Claude to its drug discovery toolkit, a deal that gives the Danish pharmaceutical group access to the recently launched Claude Science workbench to speed up R&D. The alliance, announced shortly after Anthropic's new AI workbench for researchers, aims to compress the timeline from early research to marketed medicine - a persistent pain point in an industry where bringing a single drug to market can cost billions.
The companies said they will jointly address key drug discovery challenges identified by scientists and computational teams at Novo. They plan to develop solutions for specific scientific problems and workflows that support biological reasoning. Novo has already used Claude to build NovoScribe, a tool that generates clinical study reports in minutes rather than months.
Stacking AI partnerships across the pipeline
The Anthropic deal is one of several AI bets Novo is placing as it fights to maintain market share in diabetes and obesity treatments. The company is also working with Amazon Web Services to run a London-based AI hub that uses Amazon Bio Discovery and Amazon Bedrock to identify drug targets and connect genomic, imaging, and clinical data. Separate partnerships with OpenAI and NVIDIA target internal process automation and custom generative AI models for early research.
Novo chief executive Mike Doustdar framed the push in direct terms. "AI can help us increase productivity in R&D and compress the path from research to marketed product," he said. "But beyond this, AI tools can also offer completely new scientific opportunities and aid reasoning and understanding of human biology and drug mechanics." Doustdar is currently leading a major turnaround effort at the group.
Claude's entry into pharma R&D
Under the terms of the alliance, Novo will use Anthropic models and Claude Science to "advance scientific reasoning for […] R&D and agentic software engineering." The partnership focuses on practical workflow problems that Novo's internal teams have flagged, rather than broad exploratory research. The NovoScribe tool offers an early example: clinical study reports that once took months now appear in minutes.
Anthropic CEO Dario Amodei said the collaboration targets measurable outcomes. "By giving leading researchers access to safe, capable, and trusted frontier models, we can shorten research timelines, improve outcomes, and discover medicines and treatments that significantly improve human life," he said.
For science and research professionals tracking how large language models are being applied in practice, the Novo-Anthropic deal shows a shift toward domain-specific reasoning tools. Resources like Claude AI Courses & Certifications cover the same model family now being deployed in pharmaceutical R&D settings.
Safety concerns surface alongside adoption
The pharmaceutical industry's accelerating adoption of AI arrives as technology companies issue new warnings about safety. Reports have surfaced of AI agents conversing in dialects difficult for humans to monitor, colluding to bypass guardrails, and breaking out of test environments to execute their own tasks or modify their models. These concerns sit in tension with the push to deploy frontier models deeper into drug development workflows.
Why this matters for science and research professionals
The Novo-Anthropic deal signals that major pharmaceutical companies are moving beyond pilot programs and embedding AI directly into regulated workflows like clinical study report generation. For researchers, this means AI literacy is shifting from a differentiator to a baseline expectation. Understanding how models like Claude handle scientific reasoning - and where their limits lie - will directly affect hiring, project design, and regulatory strategy in the near term. Professionals working at the intersection of AI for Science & Research should watch how the NovoScribe tool performs under real-world regulatory scrutiny, as it will set a precedent for what automated scientific writing can and cannot replace.
Your membership also unlocks:
