Archetype AI has launched Newton Agents, a portfolio of ready-to-deploy AI agents for industrial operations. The agents are powered by Newton, a foundation model trained to interpret multimodal sensor data, video, and operational context.

The portfolio includes five specific agents: Rare Event Detection, Operational State Monitoring, Anomaly Discovery, Task Verification, and Training Manual Generation. These tools are designed to detect machine failures, identify performance drivers from sensor data, and verify that manual tasks are completed correctly using video analysis.

The agents are intended to run on the cloud, at the edge, or on-premise. The company stated that the Newton model allows a single agent to generalise across an entire class of industrial assets without requiring a custom model for every individual piece of equipment. This aims to reduce the complexity of deploying AI in physical environments.

The launch follows a $35 million Series A funding round led by IAG Capital Partners and Hitachi Ventures. Investors in the company include the Amazon Industrial Innovation Fund and Samsung Ventures.

“Newton Agents change how AI is deployed in the physical world,” said Ivan Poupyrev, Co-founder and CEO of Archetype AI. He stated that the era of bespoke, one-off AI models is ending in favour of physical intelligence that can be deployed seamlessly across different environments.