British deeptech company PHYSICSX raises €255 million: the AI race expands into strategic infrastructure

The race for artificial intelligence is no longer confined to conversational models, software agents or improvements in knowledge-worker productivity. A new generation of companies is now seeking to apply AI to the physical systems underpinning the global economy. Behind PhysicsX’s €255 million funding round, led by Temasek with participation from NVIDIA, Siemens, Applied Materials, Atomico and General Catalyst, a broader contest is emerging over strategic infrastructure, industry and hardware innovation.

For several years, the debate surrounding artificial intelligence has focused primarily on the ability of models to generate text, code and images. Yet the economic value of many industries still depends on physical constraints that have lost none of their complexity. Designing an aircraft engine, optimising a semiconductor, developing a new generation of batteries or improving the energy efficiency of a data centre still requires thousands of hours of simulation, analysis and iteration.

This is precisely the problem PhysicsX intends to solve. Founded in London by former Formula 1 engineers, the company is developing an engineering platform built natively around artificial intelligence. Its objective is to accelerate the understanding and modelling of physical phenomena, dramatically reducing the time required to design and validate industrial products.

“Almost every complex problem in the physical economy, from better aircraft, chips and engines to more efficient energy systems, depends on how quickly and effectively engineers and operators can work with the underlying physics. For decades, this constraint has limited hardware innovation. AI applied to physics removes it,” says Jacomo Corbo, co-founder and CEO of PhysicsX.

The promise is ambitious. Where some simulations require several hours or even days of computing time, the company’s models aim to produce results within seconds. According to PhysicsX, this approach enables engineering teams to explore thousands of design variations where they could previously assess only a handful.

The implications extend far beyond individual productivity. In sectors such as aerospace, defence, semiconductors, energy and advanced materials, the time required to move from concept to operational product is often the principal limiting factor. As systems become more complex, development cycles are lengthening while competitive pressure intensifies.

PhysicsX believes that recent advances in model architectures, combined with the relative decline in GPU computing costs, now make it possible to deploy this approach at industrial scale. The company says its technology is already being used across aerospace, defence, semiconductors, automotive, energy and manufacturing.

PhysicsX has extended its Series B funding round, bringing the total raised to more than €133 million. The investment comes from NVentures, alongside Atomico, Temasek, Siemens, Applied Materials, July Fund, General Catalyst, NGP and other existing investors.

Beyond the funding round, another development deserves attention. PhysicsX intends to allocate part of the investment to developing a new generation of models that it calls “Large Physics Models”.

Over the past three years, the industry has become familiar with the Large Language Models powering ChatGPT, Claude and Gemini. PhysicsX is now suggesting the emergence of an equivalent category applied not to language, but to the understanding of physical phenomena.

The ambition is to train these models on vast volumes of simulations, industrial data and information from the real world, enabling them to anticipate the behaviour of complex systems. Over time, they could become foundational components of digital engineering, assisting product design in areas as varied as aircraft engines, energy reactors, advanced materials and semiconductors.

Robin Tuluie, founder and chairman of PhysicsX, believes this development could transform access to advanced engineering itself. “High-fidelity physics simulations have always been powerful, but they have remained slow, expensive and accessible only to a limited number of specialists. AI applied to physics changes that reality in every dimension.”

This drive towards democratisation represents another strategic aspect of the project. In many industrial companies, the most sophisticated simulation tools remain concentrated in the hands of expert teams. PhysicsX wants to make these capabilities available to a much broader range of users, including engineers, designers and industrial operators.

The sectors targeted by PhysicsX are equally revealing. Aerospace, defence, energy, semiconductors, advanced materials and data centres are now among the strategic priorities of the world’s leading economic powers. These industries are attracting a growing share of public and private investment related to technological sovereignty, the energy transition and the rise of artificial intelligence.

As the United States, China and Europe increase their investment in digital and industrial infrastructure, the ability to design chips, data centres, energy systems and defence equipment more rapidly is becoming a major competitive advantage. From this perspective, the next frontier of artificial intelligence may lie not in conversational interfaces, but in the software used to build the physical world.

PhysicsX’s reported growth reflects the market’s interest in this vision. The company says it has doubled recognised revenue over the past year, tripled contracted revenue and more than doubled its customer base. Its workforce now exceeds 300 people, compared with approximately half that number a year ago.

For investors, the bet is not merely on an industrial software company, but potentially on a new technological layer that could become embedded at the heart of engineering processes worldwide. After the battle over language models, strategic infrastructure may well become artificial intelligence’s next major field of expansion.