Towards a Generalizable AI for Materials Discovery: Validation through Immersion Coolant Screening

TL;DR

This paper introduces GATE, a generalizable AI framework that predicts multiple material properties simultaneously, enabling efficient discovery of immersion cooling fluids with minimal task-specific adjustments, validated through real-world screening and experiments.

Contribution

GATE is a novel AI model that learns multiple properties jointly in a shared geometric space, reducing bias and enabling transferability across different materials discovery tasks.

Findings

GATE screened billions of molecules, identifying 92,861 promising candidates.

Four candidates were experimentally validated, confirming model predictions.

GATE's performance matched or exceeded commercial coolants in tests.

Abstract

Artificial intelligence (AI) has emerged as a powerful accelerator of materials discovery, yet most existing models remain problem-specific, requiring additional data collection and retraining for each new property. Here we introduce and validate GATE (Geometrically Aligned Transfer Encoder) -- a generalizable AI framework that jointly learns 34 physicochemical properties spanning thermal, electrical, mechanical, and optical domains. By aligning these properties within a shared geometric space, GATE captures cross-property correlations that reduce disjoint-property bias -- a key factor causing false positives in multi-criteria screening. To demonstrate its generalizable utility, GATE -- without any problem-specific model reconfiguration -- applied to the discovery of immersion cooling fluids for data centers, a stringent real-world challenge defined by the Open Compute Project (OCP). Screening billions of candidates, GATE identified 92,861 molecules as promising for practical deployment. Four were experimentally or literarily validated, showing strong agreement with wet-lab measurements and performance comparable to or exceeding a commercial coolant. These results establish GATE as a generalizable AI platform readily applicable across diverse materials discovery tasks.