What PhysicsX Is and Why Its Latest Funding Round Matters
PhysicsX is a physics-based AI platform that uses machine learning and simulation models to compress complex engineering design, testing, and optimisation workflows from months of manual and compute-heavy work into seconds of automated analysis. The company’s latest PhysicsX funding round is a $300m Series C that values the business at about $2.4bn, more than double its nearly $1bn Series B valuation raised around 12 months earlier. This over-subscribed round was led by Temasek, with participation from Applied Materials, Nvidia, Atomico, General Catalyst, Siemens, M&G and Intrepid Growth Partners, among others. PhysicsX has now raised about $500m in total, signalling strong investor belief that AI engineering simulation will be a core layer of future industrial software. The capital will support platform development, AI research, and expansion, including a new office in Singapore and deeper presence in the US.
How Physics-Based AI Compresses Months of Engineering into Seconds
PhysicsX’s core promise is speed at scale: replacing slow, sequential computer-aided engineering workflows with AI models that approximate complex physics almost instantly. Traditional high-fidelity simulations for aircraft components, chips, engines or energy systems can take hours or days per configuration and require specialist engineers to set up and interpret. PhysicsX trains models on large datasets of physics simulations and real-world results so engineers can query, iterate and optimise designs through AI engineering simulation in seconds. According to PhysicsX co-founder and CEO Jacomo Corbo, “We are giving engineers the ability to explore thousands of designs where they once managed a handful, in seconds rather than weeks, across the most demanding industries in the world.” That change does more than save time: it opens up design spaces that were previously impossible to explore within project deadlines or budget limits.
A New Category: Physics-Based AI Platforms for High-Complexity Workflows
PhysicsX is positioning itself as an AI-native engineering platform, targeting the physical economy rather than office automation. Instead of focusing on email, documents or code, the company aims to automate high-complexity workflows across aerospace, defence, automotive, semiconductors, materials, energy and renewables. This physics-based AI platform sits across the engineering lifecycle: concept design, virtual testing, optimisation, manufacturing process tuning and even production operations. By embedding AI models that understand underlying physics, it gives engineers decision tools instead of replacing them, reducing the bottleneck created by scarce simulation experts and limited compute. PhysicsX calls its most advanced models “large physics models” – pre-trained systems designed to generalise across related problems and domains, much like large language models but grounded in physical laws. If they work at scale, these systems could become shared infrastructure for hardware innovation, similar to how cloud services underpinned the last software wave.
Investor Confidence and the Race to Build Large Physics Models
The over-subscribed Series C valuation of about $2.4bn reflects investor conviction that AI for engineering is a long-term platform play, not a niche tool. Temasek led the round, joined by institutional investors such as M&G Investments and Intrepid Growth Partners, alongside strategic backers including Applied Materials, Nvidia, Siemens, Atomico, General Catalyst and others. PhysicsX has doubled its team in the last 12 months to more than 300 people and plans to expand further as it builds larger, more capable large physics models. These pre-trained models are intended to give customers powerful out-of-the-box capabilities, then adapt to specific use cases with additional data. Corbo said the financing will “put that capability in the hands of more engineers and push the frontier toward ever larger and more capable large physics models,” suggesting an arms race similar to the one around general-purpose AI models, but grounded in engineering physics.
What This Means for Enterprise Engineering Workflows
PhysicsX’s rise points to a shift in enterprise engineering: physics-based AI is becoming as central to product development as CAD and traditional simulation tools. Instead of running a few expensive simulations late in a project, engineers can move to continuous, AI-driven exploration from the earliest concept stages. That means more design candidates, faster iteration cycles and better optimisation for performance, cost and sustainability. For companies facing skill shortages and tight timelines, automating large parts of simulation and virtual testing could be as transformative as early industrial automation was for manufacturing lines. Beyond speed, PhysicsX says its platform enables “more reliable, more efficient and altogether new ways of doing engineering, manufacturing and production.” If it delivers, engineering teams may spend less time setting up runs and more time on system-level decisions, with physics-based AI platforms acting as high-speed copilots across the full lifecycle of complex hardware.
