Archetype AI, the Physical AI company, launched Newton Agents, a portfolio of ready-to-deploy Physical AI agents powered by Newton, Archetype AI's foundation model for the physical world. Newton Agents understand multimodal sensor data, video, and operational context to help organizations detect machine failures before they happen, optimize processes, and verify that work gets done right. The portfolio launches with five agents that can be deployed immediately, with the flexibility to build custom agents tailored to an organization's own workflows.
Archetype AI launches Newton Agents, ready-to-deploy Physical AI agents for industrial operations.
Powered by Newton foundation model for the physical world.
Five agents: Rare Event Detection, Operational State Monitoring, Anomaly Discovery, Task Verification, Training Manual Generation.
Runs securely on cloud, edge, or on-premise.
Launches following $35M Series A led by IAG Capital Partners and Hitachi Ventures.
Investors include Bezos Expeditions, Venrock, Amazon Industrial Innovation Fund, and Samsung Ventures.
Industrial organizations have spent decades instrumenting their operations, yet the conditions that matter most (like equipment health, process stability, execution quality, emerging risk) remain difficult to understand at scale, because none can be directly measured. Rather than building a custom model for every asset and use case, Newton Agents draw on Newton's generalized physical intelligence to interpret sensor data and video out of the box, so a single agent can generalize across an entire class of assets.
The launch portfolio includes Rare Event Detection which detects rare, high-impact faults from only a handful of examples, with no massive labeled datasets required. Operational State Monitoring identifies hidden operating regimes and performance drivers from multivariate sensor time series. Anomaly Discovery surfaces emerging issues and unknown failure modes before they impact operations. Task Verification uses video to confirm tasks are completed correctly and consistently. Training Manual Generation automatically generates step-by-step SOPs and visual work instructions from operational videos.
Ivan Poupyrev, Co-founder and CEO of Archetype AI, stated: "Newton Agents change how AI is deployed in the physical world. The era of bespoke, one-off AI models is ending. With Newton Agents, organizations can deploy physical intelligence seamlessly across assets, environments, and use cases, without rebuilding every solution from the ground up."
The launch follows Archetype AI's earlier $35 million Series A, led by IAG Capital Partners and Hitachi Ventures, with participation from new and existing investors including Bezos Expeditions, Venrock, Amazon Industrial Innovation Fund, Samsung Ventures, Systemiq Capital, E12 Ventures, Higher Life Ventures, In-Q-Tel (IQT), and others. This funding supports the continued development and deployment of Physical AI solutions for industrial applications.
Newton Agents run securely wherever operations occur: cloud, edge, or on-premise. This flexibility ensures that organizations can deploy physical intelligence in the environment that best meets their security, latency, and operational requirements. Companies across manufacturing, energy, and infrastructure — including Fortune Global 500 leaders — use ready-to-deploy Newton Agents, or build their own on the Archetype Platform, to predict failures, extend equipment life, and capture expertise.
About Archetype AI
Archetype AI is the Physical AI company, helping humanity make sense of the real world. Founded by Google veterans, it created Newton, a foundation model for the physical world that learns how machines and operations actually behave directly from sensor data, adapting across many machines and sites without a custom model for each. Companies across manufacturing, energy, and infrastructure — including Fortune Global 500 leaders — use ready-to-deploy Newton Agents, or build their own on the Archetype Platform, to predict failures, extend equipment life, and capture expertise.
