Reference Architecture of SAP BTP CAP application using GPT Models

Share

As we might recall from former presentations the latest simple view of SAP Apps can be viewed as below

image

What ever we do it is designed to run on BTP (Business Technology Platform). Now that we have the Generative AI wind blowing the question is to find out what is next.

image

With the Generative AI and Deep Learning converging and creating new fronters and new Evolution comes into play.

image

Non-technical users can solve their business tasks by describing them in natural language. And foundation models are broadly applicable to many tasks without the need to re-train.

image

Reference Architecture for consuming Azure OpenAI service on SAP BTP. Details of the architecture can be found in github link.

Reference CAP application on GitHub

 

image

As a summary we will be executing on a journey below:

SAP combines the power of generative AI with the context of business data and business processes and applies fine-tuning and prompt engineering at scale. This allows SAP to turn otherwise generic results into ones that are tailored to your business. These results are built into the SAP applications that business users use every day.

SAP keeps humans in the loop responsibly to review and approve generated information.

SAP’s use of generative AI follows the same principles of business AI and the same rigorous development and responsible AI review processes.

SAP ranks #3 of 30 in Embedded AI: Omdia recognizes SAP as one of the world’s most AI-forward SaaS companies. SAP delivers a breadth of end-to-end AI capabilities built into the full context of its applications while offering customers the flexibility to deploy custom or pre-trained AI models.

image

SAP ranks #3 of 30 in Embedded AI

Get started with business AI on. See AI in action.

sap.com/ai

Share

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.