Guest Post by Rodrigo Madanes & Stefan Leichenauer
For the past few years, Large Language Models (LLMs) have rightly captured the C- suite’s attention, demonstrating impressive abilities to generate text, assist with code, and enhance customer service. But while LLMs are mastering the digital world, a new class of AI is poised to deliver a far more profound impact on the physical world and the core sectors of our economy: Large Quantitative Models (LQMs).
Unlike LLMs, which analyze existing language, LQMs are designed to process and generate quantitative data to solve complex problems in the physical world. Critically, they are anchored in the fundamental laws of nature, including physics, chemistry, and biology. This grounding in provable science gives them a decisive advantage in
applications where precision, accuracy, and reliability are non-negotiable.
Leveraging Physics-Based Simulations for Unprecedented Insight
LQMs differentiate themselves by using physics-based simulations, including advanced computational chemistry, to model quantum-level atomic and electronic behavior. This allows them to generate reliable, high-fidelity synthetic data, which is crucial for scientific research industries where acquiring real-world experimental data is often challenging.
Imagine moving from simply reading existing maps to creating new ones for uncharted territories. In biopharma (a $8 trillion market), LQMs allow drug discovery researchers to find new solutions to previously undruggable diseases by searching and optimizing in new chemical spaces. In the chemicals and materials industry (a $15 trillion market), instead of relying on slow, traditional computational methods, LQMs screen millions of potential candidates in less than 48 hours. This entails identifying innovative catalysts that enhance efficiency and supply chain resilience in crucial industrial processes, ultimately boosting production yields, reducing costs and securing a notable competitive edge.
Transforming Core Industries: Where LQMs Drive Dominant Value
The impact of LQMs, powered by these physics-based simulations, is set to revolutionize industries that form the backbone of the global economy. While LLMs demonstrate strong applicability in sectors like media creation (a $4 trillion market) and software development (a $5 trillion market), their direct applicability in transforming the
core operations of heavy industry and science-driven sectors is comparatively limited. This is where LQMs exhibit dominant potential.
Consider the financial services sector (a $10 trillion market), where companies like JPMorgan, Wells Fargo, and Citi operate. While LLMs offer value in customer interaction and document processing, LQMs are increasingly crucial for complex quantitative risk modeling activities, including fraud detection and algorithmic trading
strategies that underpin the stability and innovation in this sector.
Within the automotive and aerospace sector (a $15 trillion market), LQMs are critical for simplifying complex simulations in vehicle design. This includes aerodynamics, structural integrity analysis, allowing for the development of innovative lightweight and high-strength materials that are essential for innovation and safety.
These examples underscore that LQMs are not merely tools for incremental efficiency gains. They are powerful drivers of new value creation, enabling the development of entirely new products, materials and processes previously out of reach. This focus on creating novel capabilities and solving high-value, complex problems in the physical world is where LQMs will deliver their most significant economic impact.
Beyond a Single Model: A Modular, Reliable System for Critical Applications
LQMs should not be viewed as monolithic “black boxes” like some LLMs. They are better understood as modular “toolboxes” that can include interconnected components of physics-based equation solvers, specialized neural networks, and knowledge graphs. This architecture merges machine learning speed with verifiable scientific reliability, minimizing “hallucinations” and ensuring accuracy in critical physical-world applications.
The Path Forward: Embracing Quantitative AI for Real-World Impact
While “Quantum” is part of the “AQ” (AI and Quantum) framework guiding general technology innovation, it’s crucial for today’s leaders to understand that LQMs are delivering substantial value now. Furthermore they’re running effectively on existing classical hardware such as GPUs. These models are, architectured with the future in
mind and designed to seamlessly integrate with emerging quantum computers (QPUs), promising to further enhance their capabilities for solving certain classes of currently intractable complex problems.
For C-level executives, the message is unequivocal: to maintain a competitive edge and drive the next wave of profound innovation, a singular focus on LLMs is insufficient. Investing strategically in LQMs and the broader field of quantitative AI is essential for addressing your most pressing challenges and unlocking opportunities in the physical world. This means accelerating product development cycles, optimizing core business processes in manufacturing and energy, and creating significant new value streams across the largest sectors of the global economy.
Successfully harnessing the power of LQMs will involve building and empowering cross-functional teams that combine deep domain expertise with specialized quantitative AI capabilities. The era of quantitative AI has arrived, offering the precision, reliability and transformative potential needed to reshape industries and help solve some of the world’s most critical challenges.
The views reflected in this article are the views of the authors and do not necessarily reflect the views of the global EY organization or its member firms, or of SandboxAQ. Market sizes and company examples are illustrative based on public information and SandboxAQ analysis.
Photo From Pixels by Vishva Patel
