CAS, a division of the American Chemical Society (ACS), has announced the launch of CAS Newton℠, an agentic AI system developed for scientific discovery. CAS Newton is grounded in published scientific knowledge curated by CAS scientists and draws on more than 150 years of literature contained in the CAS Content Collection™.
According to the announcement, CAS Newton is intended to provide answers based on curated scientific knowledge to help researchers address ambiguity, conflicting results, and incomplete evidence. The release stated that, in early user feedback, three out of four respondents rated CAS Newton answers as more trustworthy than those from other AI tools. According to CAS, the system engages conversationally with complex scientific questions, retains context as questions evolve, and supports follow-up interactions that refine questions and synthesize results across multiple steps.
CAS Newton connects concepts across chemistry, biology, materials science, and intellectual property through the CAS Content Collection. It summarizes large reference sets into concise insights through agentic workflows and provides conversational access to published scientific knowledge without requiring specialized search expertise.
CAS maintains that organizations can deploy CAS Newton in secure environments and integrate it with proprietary data through Model Context Protocols (MCPs), Application Programming Interfaces (APIs), and third-party AI platforms. The company also added that the product operates within a secure application boundary, and that user input is not shared outside the solution and is not used for cross-user model training. CAS Newton is available within CAS SciFinder®, CAS BioFinder®, and through a standalone CAS Newton interface.
Click here to read the original press release.
More News in this Theme
