Radical AI

Radical AI describes a partnership with Makino to build a custom high-throughput sample preparation tool intended to compress sample-prep timelines from hours (manual/individual setup) toward a fully automated process integrated into its materials testing workflow.

Radical AI reports a head-to-head torch test where it says a Radical AI-produced alloy (RAI-939) outperformed industry-leading aerospace material C103 by 125x in environmental survivability, and it credits a collaboration with Purdue Applied Research Institute for verification.

Radical AI reports selection by AFWERX for a Direct-to-Phase II contract (amount disclosed) focused on accelerated discovery of novel high entropy alloys (HEAs) with enhanced thermomechanical properties for hypersonic-flight-related needs; the company describes its full-stack approach integrating AI-driven predictions, computational screening, and robotically operated labs.

The company announced the close of a $55M seed round (Series Seed+) to fund continued build-out of an integrated system for autonomous materials discovery using its self-driving lab workflow and AI that synthesizes literature, computational models, and experimental data.

A May 29, 2025 public submission from Joseph F. Krause (Co-Founder & CEO) discusses AI-driven materials innovation as a national investment priority aligned with the 2025 National AI R&D Strategic Plan RFI.

Radical AI announces Efficient Geometric Interatomic Potential (EGIP) and describes design choices aimed at efficiency and speed, including direct force prediction and targeted pruning; it also discusses performance on machine-learning materials benchmarks.

Radical AI releases TorchSim, describing it as a PyTorch-native atomistic simulation engine for the MLIP era and positioning it as a component of its broader “Materials Flywheel.”