Straight from
the lab.

Snowflake AI SQL User Guide: AI_SIMILARITY

Returns a semantic similarity score between two texts — understands meaning, not just character matches. Ideal for deduplication, search, and classification.

Snowflake AI SQL User Guide: AI_FILTER

Evaluate rows against a natural language condition — replace complex LIKE/regex chains with a single intent-driven prompt that returns TRUE or FALSE.

Snowflake AI SQL User Guide: AI_SUMMARIZE_AGG

Summarise an entire column of text into one result. Works with GROUP BY like COUNT() or SUM() — collapsing many rows into a single AI-generated summary.

Snowflake AI SQL User Guide: AI_AGG

Reduce a column of text into a single result using a natural language instruction. Unlike AI_SUMMARIZE_AGG, you control exactly what the output looks like.

Snowflake AI SQL User Guide: AI_TRANSLATE

Translate text between languages directly in SQL. Supports auto-detection, ISO 639-1 codes, and chains seamlessly with AI_SENTIMENT and AI_CLASSIFY.

Snowflake AI SQL User Guide: AI_SENTIMENT

Analyse sentiment — positive, negative, neutral, or mixed — across any text. Pass categories for aspect-based analysis broken down by topic.

Snowflake AI SQL User Guide: AI_REDACT

Detect and mask PII automatically — names, emails, phone numbers, SSNs and more — replacing them with category markers like [NAME] and [EMAIL].

Snowflake AI SQL User Guide: AI_CLASSIFY

Classify text into user-defined categories. Supports single-label, multi-label, label descriptions, task descriptions, and few-shot examples.

Snowflake Interval Data Type

A complete guide to all 13 INTERVAL type variations in Snowflake — year-month and day-time families, precision control, casting, and arithmetic.