German robotics firm Neura Robotics has entered a partnership with AWS to accelerate the rollout of physical AI and bring intelligent robots closer to large-scale, real-world deployment.
The collaboration will span three core areas, each designed to move physical AI from theory into practice.
First, AWS will become Neura’s primary cloud provider, hosting its “Neuraverse” platform, designed to train physical AI systems for use across robot fleets.
On the development side, Neura’s Gym training environments (where robots learn complex tasks in high-fidelity simulations) will be integrated with Amazon’s machine learning platform SageMaker.
Finally, the partners will also look at real-world deployment of Neura’s systems. As part of this effort, Amazon is planning to deploy Neura’s robots in selected fulfillment centers to gather data on how the systems operate in complex physical environments.
The deal is pitched as addressing an ongoing challenge in scaling physical AI.
Defined as systems that can perceive and reason as humans do, the tech is currently hobbled by a lack of training data. While large language models can draw on vast amounts of information from scraping the internet and other sources, robots have a much smaller foundation to learn from.
Bridging this data gap is seen as critical to unlocking AI’s full potential in industrial deployment, and the partners say they can address it by combining simulation, cloud infrastructure and operational environments.
“Physical AI will only reach its full potential if intelligence can be trained, validated, and continuously improved in the real world,” David Reger, CEO and founder of Neura, said in an emailed statement. “With AWS, we gain the infrastructure to scale the Neuraverse globally. With Amazon, we have the opportunity to bring physical AI into one of the most advanced operational environments in the world.”
“This is how physical AI moves from vision to global reality,” he added.
For Neura, the partnership is also part of a wider push to build a global robotics ecosystem, bringing together cloud providers, industrial players and semiconductor firms to support and accelerate development timelines.
Existing ecosystem members include robotics firm Kawasaki, and industrial leaders such as Schaeffler, Bosch, and Qualcomm Technologies.
