Open Platform for Enterprise AI is Welcomed by Intel

Aiming to unite the developer community for the progress of generative AI systems, Intel has welcomed the Open Platform for Enterprise AI Initiative.

The Open Platform for Enterprise AI (OPEA) was introduced as the most recent Sandbox Project by the Linux Foundation AI & Data today.

OPEA drives interoperability across a heterogeneous and diversified ecosystem, beginning with retrieval-augmented generation (RAG), with the goal of accelerating safe and economical GenAI (generative artificial intelligence) installations for enterprises.

At Intel Vision 2024, Intel CEO Pat Gelsinger outlined the industry challenges and declared the intention to create an industrywide open platform for enterprise AI.

Today is the next step in that journey and Intel is in on the ground floor.

Initially in this effort, Intel plans to:

• Publish a technical conceptual framework.
• Release reference implementations for GenAI pipelines on secure solutions based on Intel® Xeon® processors and Intel® Gaudi® AI accelerators.

• Continue to add infrastructure capacity in the Intel® Tiber™ Developer Cloud for ecosystem development, AI acceleration, and validation of RAG and future pipelines.

Through innovation and the development of new projects, the RAG ecosystem is growing. In spite of this, businesses are forced to utilize a do-it-yourself strategy since there aren’t any de facto standards among the components that enable them to select and implement RAG solutions that are open, interoperable, and accelerate time to market.

OPEA intends to address these issues by collaborating with the industry to standardize components, including frameworks, architecture blueprints and reference solutions that showcase performance, interoperability, trustworthiness and enterprise-grade readiness. This will help drive adoption of RAG solutions in the enterprise and harness innovation across an open ecosystem.