DuPont announced on November 12, 2026 a research collaboration with Microsoft and NobleAI that will broaden access to AI and machine-learning tools for its scientists and engineers, with the goal of reducing the time needed to deliver customer solutions.
Each company contributes a different component. DuPont supplies expertise in advanced solutions and applications; NobleAI provides Science-Based AI modeling through its Visualizations, Insights and Predictions (VIP) Platform; and Microsoft contributes Discovery, its enterprise agentic AI platform for R&D. DuPont’s stated goal is to help teams assess potential solutions faster, guide experiments and accelerate cycles of learning.
How the AI Tools Divide Research Work
The proposed workflow separates research orchestration from scientific modeling. Aseem Datar, Microsoft’s corporate vice president of product innovation for Microsoft Discovery & Quantum, described Discovery as an agentic environment that connects R&D teams’ data, models and know-how for scientific discovery. NobleAI CEO Alex Wang divided the software roles between Discovery’s planning and orchestration of research and VIP’s predictive and optimization modeling.
Wang described a workflow in which DuPont teams would use their existing data, expertise and processes to explore more ideas in silico, meaning through computer simulation, before concentrating physical testing on the candidates most worth pursuing. That sequence places AI earlier in candidate selection: models help narrow the options, while physical testing remains part of development rather than being displaced by simulation.
NobleAI’s modeling approach is designed for situations where industrial data is limited. The announcement describes Science-Based AI as integrating scientific principles throughout the modeling workflow, VIP as turning limited industrial data into predictive models, and NobleAI as one of Microsoft Discovery’s first launch partners. That relationship brings the modeling approach into agentic workflows for industrial R&D.
Governance and Expansion Beyond Individual Labs
DuPont is linking the technical work to a company-wide AI rollout model. Chief Information Officer Matt Abbott said, “This collaboration reflects how we are scaling AI at DuPont, with security and governance designed in from the start, so what proves out in one lab can be put to work across the company.”
That approach treats a successful laboratory result as a candidate for broader use, with security and governance forming part of the deployment design. Chief Technology Officer Marty DeGroot said the combined tools are expected to accelerate development, shorten time to market and expand innovation’s impact across DuPont’s portfolio. DuPont serves healthcare, water, construction and industrial markets, giving that portfolio-wide ambition a broad potential scope.
The practical test is whether this combination can shorten development cycles while producing results that hold up outside the initial laboratory. If it does, the governance model Abbott described would provide a route for turning a successful local workflow into a broader R&D capability across DuPont’s portfolio.