In our fast-paced and increasingly complex world, organizations are facing two fundamental challenges related to data. On the one hand, they struggle to make sense and take full advantage of the vast amounts of data they collect in their day-to-day operations. Simultaneously, in sectors like defense or emerging technology research, data is often scarce, non-existent, or error-prone. In both cases, traditional analysis methods often fall short. As a result, business leaders and government officials alike struggle to capture the timely, actionable, and trustworthy insights they need to drive their business forward, make mission critical decisions, or craft effective policies.
To bridge this gap, Galois is spearheading research into Decision Support Systems (DSS), AI/ML models designed to sift through large, ambiguous, or incomplete data sets, identifying patterns and generating insights to facilitate informed decision-making. Here, Galois advocates for the notion that machines should augment human decision-making, not replace it. While machines offer predictions, humans add meaning, context, and judgment—elements that cannot be accurately encoded into an algorithm. Our DSS are designed to maximize the respective strengths of both people and machines: using AI for precision and humans for wisdom—highlighting patterns and suggesting courses of action while leaving the ultimate judgment in human hands.
Our work spans various domains: In the defense sector, decision-support systems can assist security analysts wading through vast intelligence landscapes, sifting through vast amounts of data to prioritize threats. In healthcare, DSS could provide real-time image analysis during cancer surgeries, spotlighting areas highly probable of malignancy, empowering doctors with more informed decisions during critical operations. In the world of military and economic strategy, DSS could pair with generative AI to serve as a panel of virtual advisors—each offering a unique, data-informed perspective, but without dictating a particular course of action.
As we chart the course for future DSS, we envision systems where machines provide the tapestry of data and predictions while humans, empowered by trustworthy insights, weave the narrative of action.