SandboxAQ, a company specializing in quantum technology and artificial intelligence, announced on Tuesday that it has entered into two new agreements with the Army to assist in exploring the use of AI for discovering new materials aimed at producing faster, cheaper, longer-lasting, and more sustainable armored vehicles, electric vehicles, and drones.
The first agreement involves collaboration with the Army Futures Command’s C5ISR Center, where SandboxAQ will utilize quantitative AI software—machine learning algorithms designed to process and analyze extensive numerical and physical datasets—to develop innovative battery chemistries and designs for electric vehicles (EVs), unmanned aerial vehicles (UAVs), and portable power solutions. Jen Sovada, president of SandboxAQ’s Global Public Sector division, indicated that she could not disclose the financial details of either contract.
Sovada revealed that current batteries for EVs and similar applications often rely on rare earth minerals such as lithium, which have relatively short lifespans. Through the use of AI-generated molecular compounds, SandboxAQ and the Army aim to create more sustainable materials. “The implications of this program could enable us to develop batteries that do not depend on rare earth minerals,” Sovada explained to Breaking Defense. “This reduces our reliance on supplies from adversarial countries like China, which holds a significant share of these resources. We’re also focusing on how environmentally friendly these materials are, aiming to produce batteries that create less waste, are less toxic, and have extended shelf lives.”
The second agreement is with the Army’s Combat Capabilities Development Command (DEVCOM) Ground Vehicle Systems Center, the branch responsible for developing armored vehicles. In this partnership, SandboxAQ will collaborate with Comprehensive Carbon Impact, a company committed to addressing climate change, to leverage its AI software to identify “novel alloy materials” specifically tailored for armored vehicles, as indicated in a SandboxAQ press release.
The company asserts that these new materials could enhance the mobility, survivability, and eco-friendliness of military vehicles by reducing weight, cost, and toxicity. Sovada noted that the technology could also facilitate prototype testing simulations, significantly saving time, money, and resources—a recurring objective in the Army’s transition towards increased reliance on digital engineering processes.
Her comments likened their approach to the drug discovery process, which typically spans around 16 years. “By utilizing simulation to analyze molecular compounds and structures, we can potentially shorten the discovery and testing cycle by seven to ten years,” she said. However, while using generative AI to uncover new materials is an appealing concept, concerns exist about its practicality. For instance, Google previously reported identifying “millions of new materials” through AI, but subsequent analyses indicated that most were impractical.
To ensure the viability of the materials it develops, Sovada explained, SandboxAQ partners closely with clients, focusing on a more limited range of tailored molecules or materials that meet specific requirements. “We’re not simply generating random molecules and claiming to have created something new. Our process revolves around addressing our customers’ specific needs, allowing us to test a smaller, customized set of options,” she detailed. “Instead of overwhelming them with 200 million molecules or materials, we provide a focused selection of 10 or 50 for laboratory trials.”
If either contract yields successful results, they could support the Army’s ambitious objectives to hybridize its tactical fleet by 2035 and achieve an entirely electric fleet by 2050. Sovada expressed hope that SandboxAQ’s collaboration with the Army could pave the way for partnerships with other military branches in the future. The Army did not respond to requests for comment on Tuesday.
