Science-smart agentic AI to help you accelerate discovery with confidence.
CAS Newton is agentic AI built for scientific discovery. Carrying forward Sir Isaac Newton’s passion for connecting ideas across disciplines and decades, CAS Newton enables R&D teams to move faster and extract greater insight from over 150 years of peer-reviewed scientific data. Asking CAS Newton is like getting advice from all the most brilliant scientists in your field, past and present, through capabilities embedded directly in your workflow to spark new ideas, overcome challenges, and accelerate progress.


How CAS Newton is different
CAS Newton leverages domain-specific models and curated scientific data to provide the scientific rigor researchers expect. Unlike general AI tools, CAS Newton is science-smart, relying on trusted data from the CAS Content CollectionTM, with citations to trusted scientific literature always included.
Search summarizations on reference search results
Search Summarization in CAS SciFinder reference searches transforms dense reference sets into concise, science-aware summaries, capturing core themes and consensus findings to help you understand the landscape at a glance.
Agentic Search in CAS SciFinder® and CAS BioFinder®
Agentic Search helps you tackle multistep research workflows by conversationally orchestrating tasks, leading to query expansion, precision filtering, evidence gathering, and targeted recommendations to give you directional answers grounded in the literature.
The CAS Newton agentic search is currently in limited release for a small group of CAS SciFinder and CAS BioFinder users.
Your information remains private
Your research is your competitive edge. CAS Newton operates within a secure CAS application boundary, so no user input is shared outside the solution. It is also designed to ensure that your queries and results are not used in cross-user model training.

FAQ
What is CAS Newton?
CAS Newton is agentic AI, included within CAS SciFinder and CAS BioFinder, optimized to enable scientific discovery, and designed by scientists to understand chemistry and life sciences constructs, respect domain constraints, cite sources, and keep your work private.
Do summaries cite sources?
Yes. Summaries and query results link back to the underlying records, providing clear details on the original source information.
Is my data used to train models?
No. User queries/outputs are not used to train our models. CAS adheres to strict privacy and security controls.
How can I get access to CAS Newton?
Search summarizations that take advantage of agentic technology are already available to all users. CAS Newton chat functionality will have a phased rollout, and availability will be communicated.