OPC Foundation Enhances OPC UA for the AI Era: Companion Specifications Optimized for Agentic AI

[Cengiz Özemli]

### OPC Foundation Takes New Step for Industrial Artificial Intelligence

The OPC Foundation has announced a significant development to enable the OPC UA standard to be used more effectively in the age of artificial intelligence (AI). This initiative, initially based on the UA-for-AI-Prototype prototype work on GitHub, aims to convert more than 430 OPC UA Companion Specifications into AI-compatible, easily accessible formats.

### What are OPC UA Companion Specifications?

OPC UA Companion Specifications create consistent and machine-readable information models for specific industrial domains, devices, and use cases. They include both a written specification and a UA NodeSet file for implementation. These specifications are developed by OPC Foundation working groups and partner organizations.

### Integration with Artificial Intelligence

With this transformation, Companion Specifications content becomes available for use in search engines and AI-powered engineering workflows. By creating Markdown files, visual descriptions, RAG-compatible data chunks, and MCP and REST-based query interfaces, the use of AI in industrial automation is simplified.

### The Role of OPC UA

Companion Specifications ensure semantic interoperability across multiple vendors, products, and industries. This standardizes data not only in terms of connectivity but also in terms of content and structure, eliminating the recurring problem of data mapping in engineering solutions.

### Project Goals

The OPC Foundation aims to extend this work to all Companion Specifications, providing an accessible, consistent, and reliable semantic layer for AI systems. This is important for engineering support, semantic search, automated documentation, and the development of AI-based processes.

### OPC Foundation's Vision

"Agentic AI systems offer great opportunities in industrial automation. However, success depends on them being as accurate and reliable as existing engineering methods," says Dr. Holger Kenn, leader of the OPC Foundation AI working group. Stefan Hoppe, President of the OPC Foundation, emphasizes that with the widespread use of these standards, industrial data will be utilized more effectively.

### Technical Specifications

• More than 430 OPC UA Companion Specifications are being made AI-ready
• Conversion formats: RAG (Retrieval-Augmented Generation), MCP (Model Context Protocol), REST query interfaces
• AI-ready file formats: Markdown, token-optimized data chunks, vector embeddings
• Industrial information standardization and semantic interoperability are provided

### Conclusion

The OPC Foundation continues to align OPC UA-based information models with artificial intelligence to support the reliable and effective use of AI in the industrial sector. These developments are critical for the widespread adoption of AI-powered applications in the manufacturing and automation sectors.