Anthropic says its public benefit mission includes accelerating the rate of scientific progress by building AI systems that assist — and eventually autonomously contribute to — research discovery.

Until recently, scientists typically used Claude for discrete tasks like writing statistical code or summarizing academic papers. Now, the company’s ambition is to make Claude a partner across the entire research lifecycle, from hypothesis generation to commercial distribution.

“Increasing the rate of scientific progress is core to our mission,” Anthropic stated in its announcement. “We’re focused on building the tools to allow researchers to make new discoveries – and eventually, to allow AI models to make these discoveries autonomously.”

At the core of the upgrade is Claude Sonnet 4.5, Anthropic’s most advanced model to date. It significantly outperforms prior versions on life sciences benchmarks.

On Protocol QA, a benchmark that tests an AI model’s understanding and reasoning across laboratory procedures and experimental protocols, Claude Sonnet 4.5 achieved a 0.83 score, surpassing the human baseline of 0.79 and the 0.74 scored by Sonnet 4.

On BixBench, which measures bioinformatics task performance, it demonstrated similar gains, showing improved comprehension and accuracy.

These advancements, Anthropic says, make Claude a more reliable collaborator for complex laboratory, clinical, and regulatory work.

Anthropic is expanding Claude’s ecosystem with several new connectors to scientific platforms, the ability to use Agent Skills, and life sciences–specific support in the form of a prompt library and dedicated expert assistance. These updates directly integrate Claude with leading research tools, giving scientists seamless access to systems they already use:

These new scientific connectors build on Claude’s existing integrations with enterprise and data platforms, expanding its utility for bioinformatics, collaborative research, and enterprise-scale analytics:

Together, these integrations allow researchers and enterprise teams to embed Claude’s reasoning capabilities directly into their data pipelines and collaboration tools, creating a more unified environment for scientific discovery and analysis.

Anthropic also introduced Agent Skills — structured folders of instructions, scripts, and resources that improve Claude’s ability to perform repeatable tasks.

The company is creating its own suite of scientific skills, beginning with single-cell-rna-qc, which automates quality control and filtering for single-cell RNA sequencing data using scverse best practices.

Scientists can also build custom skills, allowing them to encode their own lab protocols and data workflows directly into Claude’s memory. Anthropic provides detailed documentation for developing and deploying these custom skills.

Anthropic outlines a broad set of use cases where Claude can assist scientific and commercial teams:

To accelerate adoption, Anthropic is developing a prompt library tailored to common scientific tasks, helping researchers get precise results faster.

Anthropic is collaborating with consulting and technology leaders to bring AI-driven discovery into mainstream life sciences workflows. Partners include Caylent, Deloitte, Accenture, KPMG, PwC, Quantium, Slalom, Tribe AI, and Turing, alongside AWS and Google Cloud.

Sanofi, a leading pharmaceutical company, is already deploying Claude internally.

“Claude, paired with internal knowledge libraries, is integral to Sanofi’s AI transformation and used by most Sanofians daily in our concierge app,” said Emmanuel Frenehard, Chief Digital Officer at Sanofi. “We’re seeing efficiency gains across the value chain, while our enterprise deployment has enhanced how teams work. This collaboration with Anthropic augments human expertise to deliver life-changing medicines faster to patients worldwide.”

Beyond commercial offerings, Anthropic continues to back fundamental research through its AI for Science program. The initiative provides free API credits to leading labs and researchers tackling high-impact scientific problems.

These collaborations not only accelerate academic progress but also help Anthropic identify new opportunities for applying Claude’s reasoning capabilities in the scientific domain. The company continues to invite submissions for new project ideas through its AI for Science program, encouraging researchers to propose high-impact studies that leverage Claude’s tools for discovery.

Jonah Cool and Eric Kauderer-Abrams, who lead Life Sciences Partnerships and R&D at Anthropic, discuss Claude for Life Sciences and other recent initiatives in this overview video below — offering deeper insight into how the company is advancing AI-assisted research.

Claude for Life Sciences is now available on Claude.com and the AWS Marketplace, with Google Cloud Marketplace availability coming soon.

Researchers and organizations can request live demos or sign up for early integration support directly through Anthropic.

Q1: What is Claude for Life Sciences?
A: It’s an enhanced version of Anthropic’s Claude AI optimized for biomedical and research workflows, from literature review to regulatory compliance.

Q2: What makes Claude Sonnet 4.5 different?
A: It delivers higher accuracy on scientific benchmarks like Protocol QA and BixBench, exceeding even human baseline performance.

Q3: What are “connectors” and how do they help?
A: Connectors let Claude directly interact with scientific platforms like Benchling, BioRender, and PubMed for contextual, source-linked outputs.

Q4: How do Agent Skills improve performance?
A: Agent Skills teach Claude to follow precise scientific procedures, ensuring consistency and protocol adherence across tasks.

Q5: How can scientists get access?
A: Claude for Life Sciences is accessible via Claude.com and AWS Marketplace, with Google Cloud support coming soon.

Anthropic’s Claude for Life Sciences signals a major shift in how AI systems can contribute to real-world scientific discovery. By embedding Claude directly into lab and research workflows, Anthropic is bridging a critical gap between computational reasoning and experimental validation — a challenge that has historically slowed progress across the biomedical field.

The potential impact extends well beyond productivity. As life sciences grow increasingly complex, AI models like Claude Sonnet 4.5 can help researchers navigate massive data sets, interpret ambiguous results, and identify promising avenues for new treatments or technologies. This could dramatically accelerate drug development, genomic research, and biotech innovation, while reducing the time and cost needed to bring discoveries from concept to clinic.

Equally important, Claude’s transparent integrations and skills-based architecture offer a responsible blueprint for scientific AI — one where humans remain in control of design and oversight, while machines handle scale and the complexity of data analysis.

In practical terms, this evolution positions Anthropic not just as a model developer but as a partner in AI-augmented science, reshaping how labs, pharmaceutical companies, and research institutions approach discovery.

If the system performs as intended, Claude for Life Sciences could become a foundational layer of the next generation of AI-driven R&D infrastructure, helping scientists answer deeper questions faster — and ultimately improving global health outcomes through accelerated innovation.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant used for research and drafting. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.