Claude Science bets on workflow over models for labs
anthropic launched claude science, a vertical workbench using existing models to unify scientific tools, signaling a shift from raw model capability to owning the operating layer for specific industries.
anthropic introduced claude science, a dedicated workbench for computational research that connects over 60 scientific databases and uses multi-agent workflows to delegate tasks without requiring new model training. the system includes a fact-checker agent to verify citations and calculations, addressing reproducibility issues in ai-assisted scientific writing by capturing the exact code and history for every figure. users can run the tool on their own infrastructure, preserving data privacy while using plain language prompts to edit code and visualizations. anthropic's approach contrasts with openai's specialized gpt-roslind model and google deepmind's proprietary foundation models, positioning claude as an operating layer that uses existing models to streamline scientific workflows.