A data engineering team is implementing a solution using Snowflake Cortex's AI_COMPLETE function to process customer support tickets. They are concerned about sensitive information and ensuring the model's responses are safe, while adhering to Snowflake's data governance principles. Which of the following statements correctly describe the functionality of Cortex Guard and Snowflake's data privacy commitments in this context?

• A. Option D
• B. Option E
• C. Option A
• D. Option B
• E. Option C

A Snowflake developer is tasked with enhancing a daily data pipeline. The pipeline processes raw text descriptions of system events and needs to extract structured information, specifically the (string) and its (string, restricted to 'low', 'medium', 'high', 'critical'). The output must be a strictly formatted JSON object, ensuring data quality for downstream analytics.
Consider the following SQL snippet intended for this transformation:

Which of the following statements are correct regarding this implementation and best practices for using with structured outputs in a data pipeline?

• A. Using 'TRY COMPLETE instead of would allow the pipeline to gracefully handle cases where the model fails to generate a valid JSON response by returning 'NULL' instead of an error.
• B. The complexity of the JSON schema, particularly deep nesting, does not impact the number of tokens processed and billed for 'AI_COMPLETE Structured Outputs.
• C. Setting 'temperature 'to '0.7 ' is optimal for ensuring the most consistent and deterministic JSON outputs, especially for complex extraction tasks.
• D. For all models supported by 'AI_COMPLETE' Structured Outputs, the 'additionalPropertieS field must be set to 'false' in every node of the schema, and the 'required' field must contain all property names.
• E. The 'response_format' correctly defines the expected JSON structure, using 'enum' for 'severity_lever and 'required' to ensure 'event_name' and severity_lever are always present if extracted.

An organization is building a new knowledge base system within Snowflake, which relies on 'SNOWFLAKE.CORTEX.EMBED_TEXT_1024' to generate and store embeddings for documents in a 'VECTOR(FLOAT, 1024)' column. They plan to use these embeddings for semantic search and integrate them into various data processing workflows. Which of the following statements accurately describe limitations or specific compatibility aspects of 'EMBED TEXT 1024' or the 'VECTOR' data type within Snowflake?

• A. The 'VECTOR data type, used to store the output of is fully supported as primary or secondary index keys in Snowflake's hybrid tables.
• B. The 'EMBED function in the Cortex REST API can be used to process a list of text strings, where each individual string can be up to 4096 characters long.
• C. The 'VECTOR data type is not supported in 'VARIANT columns, which means direct storage of embeddings alongside other semi-structured metadata in a single "VARIANT column is not possible.
• D. When 'EMBED_TEXT 1024' is invoked within a Snowflake dynamic table's SELECT statement, it allows for continuous, automated updates of embeddings as new data arrives.
• E. To compare the generated 1024-dimension vectors for similarity, only the 'VECTOR COSINE SIMILARITY function is officially supported by Snowpark Python.

A security administrator is implementing strict model access controls for Snowflake Cortex LLM functions, including those accessed via the Cortex REST API. By default, the 'SNOWFLAKE.CORTEX USER' database role is granted to the 'PUBLIC' role, allowing all users to call Cortex AI functions. To enforce a more restrictive access policy, the administrator revokes 'SNOWFLAKE.CORTEX USER from 'PUBLIC'. Which of the following actions must the administrator take to ensure specific roles can 'still' make Cortex REST API requests, and what are the implications?

• A. The from 'SNOWFLAKCORTEX USER database role is only required for SQL functions, not for the Cortex REST API, so no further action is needed after revoking 'PUBLIC for REST API access.
• B. The 'SNOWFLAKE.CORTEX USER database role must be granted directly to individual users who need access, as it cannot be granted to other account roles.
• C. The 'SNOWFLAKE.CORTEX USER database role must be granted to the specific account roles, and then these account roles must be granted to users. Additionally, the account parameter can be used to restrict which models are accessible.
• D. Access for Cortex REST API is managed independently of database roles; a separate REST API key must be provisioned for each user or application.
• E. Only the role can make cortex REST API calls after revoking 'SNOWFLAKE.CORTEX_USER from 'PUBLIC', as this role inherently bypasses all other access controls.

A business intelligence team wants to enable non-technical users to query structured data in Snowflake using natural language. They are considering Cortex Analyst. What is the primary role of a semantic model in Cortex Analyst to achieve this goal for structured/text-to-SQL use cases?

• A. The semantic model provides a mapping between business-friendly terms and the underlying technical database schema, enhancing the LLM's ability to generate accurate SQL from natural language questions.
• B. The semantic model acts as a vector store, storing embeddings of all data columns to enable semantic search for natural language queries.
• C. The semantic model directly executes SQL queries provided by end-users, bypassing the need for an LLM to generate them.
• D. It stores user authentication credentials and data access policies, ensuring that only authorized users can interact with the data.
• E. It serves as a cache for frequently requested data, reducing latency for natural language queries by providing pre-computed results.