Google DeepMind Launches European Robotics Accelerator

Google DeepMind has kicked off its Accelerator: Robotics program in London, bringing together 15 early-stage startups from across Europe for three months of technical mentorship, product guidance, and access to Google's AI stack and Gemini robotics models.

The company says the selected founders, drawn from what it describes as a strong pool of applicants, convened this week to meet Google DeepMind and Google teams and begin the program. The cohort spans 11 countries, from Norway to Greece, and covers applications ranging from logistics and manufacturing to healthcare, climate, and advanced navigation.

The program's stated goal is to help founders turn cutting-edge AI research into real-world robotics applications. "Robotics is one of the most exciting frontiers of AI, where advances in language, vision and action models can help create intelligent machines that interact with the real world in safer, more helpful and more adaptive ways," Google wrote in its announcement.

The bet matters beyond the startup demo circuit. Europe has struggled to translate its research strength in machine learning into commercial AI leaders, and physical AI — robots that perceive, reason, and act in the real world — remains a contested frontier where capital, talent, and compute concentration largely sit with US and Chinese players. By attaching early-stage European robotics companies to its models and infrastructure, Google DeepMind positions itself as a platform layer for the region's embodied AI ecosystem.

The cohort

The 15 startups reflect the breadth of embodied AI applications now reaching commercial maturity:

3D-Components AS (Norway) is developing RobTrack, an AI-driven platform that automates parameter selection and quality control for robotic welding and metal 3D-printing. The company claims it runs 280X faster than current practices.

Acumino (Greece) builds hardware-agnostic physical AI that enables robots to perform complex industrial tasks in a scalable, cost-efficient, and reliable manner with high ROI.

Adapta Robotics (Romania) deploys physical AI that replicates human touch to test devices and software for healthcare, automotive, and consumer electronics, targeting automated QA and the circular economy.

AUAR (United Kingdom) makes homebuilding more affordable by deploying robotic MicroFactories directly to construction sites.

Bubble Robotics (France) is building what it calls the ocean's autonomous workforce: a vessel-free constellation of self-docking surface and subsea robots that feed a live underwater world model.

Danu Robotics (United Kingdom) uses embodied AI robotic systems to automate complex waste sorting, increasing efficiency, improving safety, and enabling scalable recovery of valuable materials.

Deltia GmbH (Germany) digitizes production-line work, transforming workflows into process graphs that help teams optimize manual processes and automate repetitive tasks.

Embodied AI (Switzerland) deploys teleoperated humanoids that collect data during customer service to continuously train and improve their manipulation skills.

Extend Robotics (United Kingdom) provides teleoperation software and data pipelines that help train and fine-tune foundation models for real-world robotics applications.

Forgis (Switzerland) develops AI agents that understand machines like experienced engineers, predicting failures and optimizing operations.

Generative Bionics (Italy) develops humanoid robots based on physical AI, built in Europe but designed to scale globally.

Qualia (Denmark) is building infrastructure that enables companies to turn robotic foundation models into working deployments, automating and optimizing time-consuming manual labor.

ROBEAUTE (France) is building microrobots that navigate through brain tissue to diagnose, treat, and monitor neuropathology — a new physical infrastructure layer in neurosurgery.

Staer (Sweden) uses computer vision on existing cameras and sensors to build 3D spatial models of facilities, giving robots a shared environment to navigate and operators real-time visibility into operations.

Touchlab (United Kingdom) uses advanced nano inks to create an "e-skin" that gives robots a high-resolution sense of touch across flexible surfaces.

Why it matters

Two patterns stand out in the cohort. First, teleoperation and data collection dominate: Embodied AI and Extend Robotics both focus on pipelines that gather real-world manipulation data to train foundation models — a signal that data acquisition, not just model architecture, is a bottleneck in robotics. Second, several startups — Qualia, Staer, Forgis — are building the middleware and deployment infrastructure layer, betting that turning foundation models into reliable deployments is itself a business.

Access to Google's Gemini robotics models gives these companies a shortcut through that bottleneck. It also deepens their dependence on Google's stack, a trade-off European founders have weighed repeatedly as the region debates technological sovereignty in AI.

Each company will receive mentorship and strategic guidance from Google DeepMind and Google, according to the announcement, with the stated aim of helping them accelerate development and scale responsibly. The program runs for three months in London. Google DeepMind points interested founders to the official program page — a clear signal it intends to run this as a recurring pipeline into Europe's robotics ecosystem rather than a one-off.