Franz Tschimben (ALLSIDES) & Michael Brehm & Ben Scheidt (Redstone): Why physical AI needs 3D data

The next wave of AI will not be driven by larger general-purpose models alone. It will be verticalized, grounded in the physical world, and trained on reality-grade 3D data.In this EUVC episode, host Andreas Munk Holm speaks with Michael Brehm (General Partner and Founder at Redstone), Ben Scheidt (Partner at Redstone) and Franz Tschimben (Co-Founder of ALLSIDES).They explore why one-size-fits-all AI falls short in robotics, simulation, and other physical applications, and how ALLSIDES converts physical objects into relightable 3D digital twins in minutes.ALLSIDES has worked with clients such as Meta, Amazon, Nike, adidas, Zalando, NUREG, GORE-TEX, La Sportiva, and Inditex Group brands including Zara, Massimo Dutti, and Bershka. It also has a deep integration with NVIDIA.Key takeawaysAs foundation models commoditize, defensibility shifts toward specialized data and infrastructurePhysical AI and world models require high-fidelity 3D data, not only internet-derived 2D contentReality-grade digital twins at scale can unlock robotics, simulation, generative 3D, digital commerce, and AI content creationBuilding global deep tech from Europe requires proximity to customers and demand in the US and AsiaTimestamps(00:00) The 3D data opportunity(01:05) Why AI needs real-world grounding(02:20) Why AI is becoming verticalized(05:10) Building the data layer for physical AI(07:25) Why robots need 3D data(09:10) Automated 3D scanning at scale(13:15) Market size and the platform opportunity(17:00) Mapping the physical world(21:00) Customers, NVIDIA and market adoption(26:20) Building a global deep-tech companySubscribe to EUVC for more insights.