Anthropic launched Claude Science on June 30, 2026 — a dedicated AI workbench for scientific researchers that integrates more than 60 curated databases and specialist toolkits. Alongside the product launch, the company announced it is starting its own internal drug discovery program focused on neglected diseases, marking a bold pivot from AI toolmaker to active participant in biomedical research.
Key Highlights
- Claude Science integrates 60+ scientific databases including UniProt, PDB, Ensembl, ChEMBL, and ClinVar
- A coordinating agent manages multi-step workflows while specialist sub-agents handle genomics, proteomics, structural biology, and cheminformatics
- A dedicated reviewer agent checks citations and calculations before every output
- Early users report research tasks completing in roughly one-tenth of previous timelines
- Available in beta to Pro, Max, Team, and Enterprise subscribers
- Applications for up to $30,000 in research credits are open through July 15, 2026
A Workbench, Not Just a Chat Interface
Claude Science departs from general-purpose AI assistants by running as a local application on macOS or Linux, or remotely via SSH and HPC login nodes. Large datasets stay within a laboratory's own infrastructure — only the necessary context is sent to Claude. Every output includes the full code, environment specifications, and message history, making results fully reproducible and auditable.
The workbench natively renders 3D protein structures, genome browser tracks, and chemical structures directly in the interface. It also connects to BioNeMo models including Evo 2, Boltz-2, and OpenFold3, giving researchers access to state-of-the-art biological foundation models alongside the conversational AI layer.
Early Results From Research Partners
Anthropic has already partnered with institutions including the Allen Institute, UCSF Brain Tumor Center, Manifold Bio, and Novo Nordisk. The results reported so far are striking.
Jérôme Lecoq at the Allen Institute described Claude Science reducing grant review writing from roughly two years of accumulated effort to an accelerated timeline across more than 100 extensive review documents. Stephen Francis at the UCSF Brain Tumor Center said analytical workflows that previously took a full research cycle were completed in approximately one-tenth of the time.
Independent academic researchers cited estimates that AI-assisted drug development pipelines could compress timelines from the traditional 10-to-15-year window down to 2-to-5 years.
Anthropic's Internal Drug Discovery Bet
Beyond the product, Anthropic is establishing its own pre-clinical drug development program. Eric Kauderer-Abrams, the company's head of life sciences, explained the rationale directly: "We're doing this because we believe first and foremost that to build the right models, products and tools to accelerate the industry, we need to live it along with all of you."
The program will focus on neglected diseases — conditions that traditional biopharmaceutical companies typically pass over because the commercial return does not justify development costs. The move also signals a strategic choice to avoid competing directly with pharma customers on their own therapeutic pipelines.
The hire of John Jumper — the Nobel Prize co-recipient behind AlphaFold at Google DeepMind — as a senior research scientist at Anthropic earlier this year was widely read as a signal of the company's scientific ambitions long before the Claude Science announcement.
Pricing and Access
Claude Science is available in beta to all Pro, Max, Team, and Enterprise subscribers. Academic and non-profit research laboratories can access discounted Team plan pricing. Anthropic is funding up to 50 research projects with grants of up to $30,000 in model credits each, with applications closing July 15, 2026. Modal is providing an additional $2,000 in compute credits per accepted project. Notification of accepted projects is expected by July 31, 2026, with project periods running from September 1 through December 1, 2026.
What's Next
Claude Science represents a structural shift in how Anthropic positions itself in the AI market. By combining a product aimed at pharmaceutical and academic researchers with an in-house drug discovery program, the company is betting that the most credible way to build scientific AI is to conduct science itself. Whether that hypothesis holds will depend on what Anthropic publishes from its own labs in the months ahead.
Source: Anthropic