Amazon is heavily investing in technology focused on understanding our physical world through what is known as Physical AI. This approach differs from traditional AI, which primarily generates text or images, by emphasizing real-time perception and decision-making based on real-world inputs. At the heart of Amazon's strategy is a startup named Archetype AI, which secured a $35 million Series A funding round in November 2025, backed by Amazon’s Industrial Innovation Fund.
#What does Archetype AI do that’s different?
Archetype AI has developed an innovative model called Newton. Unlike conventional AI that processes vast amounts of text from the internet, Newton is trained on the chaotic data sourced from physical environments. It analyzes multimodal sensor inputs—text, audio, and video—enabling it to understand and interact with its surroundings in real-time. This capability is seamlessly integrated with Amazon Web Services (AWS), allowing any business operating on Amazon’s cloud infrastructure to leverage its unique features, enhancing operational efficiencies.
#How broad is Amazon's strategy in Physical AI?
Archetype AI is just one of the initiatives Amazon is pursuing in the realm of Physical AI. In September 2025, Amazon collaborated with NVIDIA and MassRobotics to establish the Physical AI Fellowship, aimed at supporting startups dedicated to simulation-based robotics and automation. This program is designed to foster innovation by providing significant resources, mentorship, and essential computing infrastructure. The fellowship is currently accepting applications for the 2026 cohort.
The collaboration with NVIDIA utilizes the Isaac Sim platform, which runs on AWS. This environment allows companies to design and train robots within simulated settings, reducing risks and costs before they are deployed in real-world scenarios. Reports indicate that projects leveraging this platform have achieved a 25% efficiency improvement within Amazon’s own supply chain operations.
#Why is simulation technology essential for Physical AI?
The importance of simulation technology lies in its ability to offer a safe space for companies to test and refine their models. By simulating realistic scenarios, businesses can gather ample data and train their AI models without the financial burdens associated with physical trials. Only after proving their effectiveness in simulations do these models transition to practical applications.
In addition to Amazon’s direct initiatives, other projects like Prometheus are also actively developing Physical AI technologies independently, focusing on engineering applications in the physical world.
#What should investors be aware of?
Amazon's strategy of investing through its Industrial Innovation Fund, hosting startups on the AWS Marketplace, and co-sponsoring fellowships forms a self-reinforcing cycle of development in Physical AI. Investors should closely monitor how the Newton model from Archetype AI performs on the AWS Marketplace and note whether the upcoming Physical AI Fellowship cohort yields startups capable of securing significant follow-on funding. These developments could signal the next wave of profitability and innovation in the tech industry.