Redwood AI Announces Optimization Module Update to Reactosphere, Expanding Experimental Planning and Chemical Process Optimization Capabilities

Rhea-AI Summary

Redwood AI (OTCQB:RDWCF) announced an update to its Reactosphere platform with a new Optimization Module for AI-guided chemistry workflows. The module combines Bayesian optimization, experimental design, and proprietary sample-size planning to guide reaction condition selection, improve yield and purity, and reduce unnecessary experiments across pharmaceutical, materials, specialty chemical, and defense-related applications.

AI-generated analysis. Not financial advice.

VANCOUVER, BC / ACCESS Newswire / May 14, 2026 / Redwood AI Corp. (CSE:AIRX)(OTCQB:RDWCF)(Frankfurt:Y0N, WKN: A422EZ) ("Redwood" or the "Company") is pleased to announce an expansion of the optimization capabilities within Reactosphere (the "Platform" or the "Software"), its AI-powered chemistry platform, through the launch of a new Optimization Module (the "Module"). The Module is designed to help chemists and R&D teams improve experimental outcomes while reducing the time, material usage, and trial-and-error typically associated with chemical optimization workflows.

The Optimization Module expands Reactosphere beyond reaction planning and sourcing intelligence by introducing guided experimental optimization workflows that combine Bayesian optimization¹, experimental design, and sample-size planning into a unified system. The Company believes this enhancement will help users improve reaction yield, purity, and process efficiency while reducing unnecessary experimentation and improving decision-making across development programs.

Chemical optimization often requires multiple experimental rounds across complex variable spaces, including reaction conditions, catalysts, solvents, and reagent concentrations. The Optimization Module is designed to support this process by recommending subsequent experimental conditions based on prior results through either fully sequential or batch-sequential workflows. The Module also incorporates multiple acquisition strategies, allowing users to balance immediate performance improvement, broader exploration of experimental space, and uncertainty reduction depending on program objectives.

To support experimental planning before laboratory work begins, the Module introduces Redwood's proprietary sample-size planning system, designed to estimate the number of experiments required to achieve a target level of predictive accuracy. Combined with structured initial experimental design generation and support for both numeric and categorical variables, Redwood believes this capability can improve early-stage data quality, strengthen downstream model performance, and help users optimize experimental resources more efficiently.

Redwood believes these enhancements further strengthen Reactosphere's position as an integrated chemistry intelligence platform, supporting not only reaction prediction and chemical sourcing, but also experimental execution and optimization. The Company expects the Module may be applicable across pharmaceutical development, materials science, specialty chemicals, and defense-related chemistry applications where efficient experimental decision-making is critical.

"Optimization is one of the most iterative and resource-intensive parts of chemistry development. By introducing structured experimental planning and AI-guided optimization into Reactosphere, we believe we can help teams improve how they design experiments, reduce unnecessary experimental burden, and reach stronger outcomes more efficiently. This expands the practical utility of the Platform and further supports its use across commercial, industrial, and defense-related chemistry applications," said Louis Dron, CEO of Redwood AI.

About Redwood AI Corp.

Redwood AI uses advanced artificial intelligence to accelerate chemistry R&D, with the aim of assisting in drug discovery and development, and furthering defense and safety solutions. The Company combines expertise in chemistry, AI, and manufacturing to streamline drug synthesis and scale-up. Redwood AI's platform is designed to enable faster, more efficient development of new therapies and chemistry-driven applications.

ON BEHALF OF REDWOOD AI CORP.,

"Louis Dron"
Chief Executive Officer

For more information, please contact:

Louis Dron
Chief Executive Officer
Tel: +1 888 530 8488
investors@redwoodai.com

The CSE and Information Service Provider have not reviewed and do not accept responsibility for the accuracy or adequacy of this release.

Forward-Looking Statements Caution. This news release contains statements and information that, to the extent they are not historical fact, may constitute "forward-looking information" within the meaning of applicable securities legislation. Forward-looking information is generally identified by words such as "believe", "expect", "anticipate", "estimate", "intend", "plan", "may", "should", "will", "potential" and similar expressions and, in this news release, includes statements relating to the development and potential deployment of the AI-powered Platform, and the expectation that the Software may be utilized for drug discovery or development or to further defense or safety solutions. Although the Company believes that the expectations and assumptions on which such forward-looking information is based are reasonable, undue reliance should not be placed on it, as actual results may differ materially from those expressed or implied. Forward-looking information inherently involves risks and uncertainties, many of which are beyond the Company's control. The forward-looking information contained in this news release is made as of the date hereof, and the Company undertakes no obligation to publicly update or revise such information, whether as a result of new information, future events or otherwise, except as required by applicable laws.

¹ Bayesian optimization is a sample-efficient optimization method used to find improved outcomes when experiments or evaluations are expensive or time-consuming. It uses a predictive model and an acquisition function to help decide which experiment or condition to test next. See On Local Optimizers of Acquisition Functions in Bayesian Optimization https://arxiv.org/abs/1901.08350.

SOURCE: Redwood AI Corp.

View the original press release on ACCESS Newswire

FAQ

What did Redwood AI (OTCQB:RDWCF) announce on May 14, 2026 regarding Reactosphere?

Redwood AI announced a new Optimization Module for its Reactosphere chemistry platform to enhance experimental planning and process optimization. According to Redwood AI, the module adds guided optimization workflows that combine Bayesian optimization, experimental design, and sample-size planning in a unified system for chemistry R&D teams.

How does Redwood AI's Reactosphere Optimization Module improve experimental planning for chemists?

The Optimization Module supports chemists by recommending subsequent experimental conditions based on prior results in sequential workflows. According to Redwood AI, it integrates sample-size planning, structured initial design generation, and support for numeric and categorical variables to estimate experiment counts and improve early-stage data quality and model performance.

What optimization workflows are supported by Redwood AI's new Optimization Module in Reactosphere?

The Optimization Module supports fully sequential and batch-sequential experimental optimization workflows driven by prior experimental results. According to Redwood AI, it also offers multiple acquisition strategies so users can prioritize immediate performance gains, broader exploration of experimental space, or uncertainty reduction, depending on specific program objectives and constraints.

How does Redwood AI's sample-size planning system in Reactosphere help chemical R&D teams?

Redwood AI's proprietary sample-size planning system estimates how many experiments are needed to reach a desired predictive accuracy. According to Redwood AI, this capability, combined with structured initial experimental design, aims to enhance early data quality, strengthen downstream models, and optimize experimental resource allocation.

Which industries could benefit from Redwood AI's Reactosphere Optimization Module (RDWCF)?

The Optimization Module may be applicable to pharmaceutical development, materials science, specialty chemicals, and defense-related chemistry. According to Redwood AI, it is designed for settings where efficient experimental decision-making, improved reaction yields, higher purity, and reduced experimental burden are important across commercial, industrial, and defense-focused chemistry programs.

What specific capabilities does Redwood AI's Optimization Module add beyond reaction planning and sourcing intelligence?

The module extends Reactosphere with guided optimization workflows, Bayesian optimization, and integrated sample-size planning tools. According to Redwood AI, these additions support experimental execution and optimization, helping users refine reaction conditions, improve process efficiency, and reduce unnecessary trial-and-error in complex chemical development workflows.