Autonomous AI Database Select AI

The image shows how Autonomous AI Database Select AI works. The diagram outlines how you can have a conversation with your data by asking a natural language question through an interface, such as an integrated development environment or application, via text or voice.

Autonomous AI Database Select AI then uses a large language model (LLM) to generate a SQL query and performs the following tasks:

• 1) Augments the natural language question with metadata from the schema(s) identified in the user’s profile.
• 2) Feeds the LLM with an augmented prompt.
• 3) LLM produces a SQL query against the database.
• 4) Query is run and the result is sent to the user.
• 5) Previous questions are retained for conversation-like user interactions.

Finally, the user receives a response back with the answer from their own organization’s data, based on its existing data security policy.

This image shows how Autonomous AI Database Select AI works. The diagram shows how you generate personalized content just by asking a question into the Select AI prompt.

A user kicks off a workflow through an application. For example, the user wants to create a promotional offer based on a customer’s previous purchases.

The application utilizes data in Autonomous AI Database and creates personalized targeted promotional offers via a large language model (LLM) including these steps:

• 1) Context from Autonomous AI Database (for example, customer demographic info and purchasing behavior, products that need to be promoted, etc.) is queried.
• 2) Prompt task instructions are combined with this data (for example, recommend similar products from the promoted product list; write a personalized and convincing e-mail with the recommendations)
• 3) It then feeds the LLM with the augmented prompt and processes the result.

The final output is provided to the user that includes a compelling promotional email offer with personalized product recommendations based on customer information, behavior and past purchases.

The image shows how Autonomous AI Database Select AI works. The diagram outlines how you can build automate the creation and population of vector store from text files such as, txt and html, on your object store.

Select AI automatically processes documents to chunks, generates embeddings, stores them in the specified vector store, and updates the vector index as new data arrives:

• 1) Input: Data is initially stored in an object storage.
• 2) Autonomous AI Database retrieves the input data or the document, chunks it, and passes to an embedding model.
• 3) The embedding model processes the chunk data and returns vector embeddings.
• 4) Output: The vector embeddings are stored in Autonomous AI Database as a vector data type for use with RAG. As content is added, the vector index is automatically updated.

The image shows how Autonomous AI Database Select AI works. The diagram outlines how Select AI implements retrieval-augmented generation (RAG).

RAG retrieves relevant pieces of information from the enterprise database to answer a user's question. This information is provided to the specified large language model along with the user prompt. Select AI uses this additional enterprise information to enhance the prompt, improving the LLM's response. RAG can enhance response quality with update-to-date enterprise information from the vector store:

• 1) Input: User asks a question.
• 2) Autonomous AI Database Select AI generates vector embeddings of the prompt using the embedding model specified in the AI profile.
• 3) Autonomous AI Database Select AI uses the generated embedding and AI Vector Search to find similar content from the customer’s enterprise data.
• 4) The vector search returns the top K results which will be used to augment the prompt.
• 5) Autonomous AI Database sends the top K query results with user question to the LLM.
• 6) The LLM returns its response to the Autonomous AI Database instance.
• 7) Output: Autonomous AI Database Select AI provides the response to the user