Amazon Quick adds autonomous agents and multi-dataset analytics

Full analysis

AWS has announced several new features for Amazon Quick, its AI assistant for business applications. The headline addition is autonomous agents: users describe tasks in natural language and set "autonomy levels" ranging from step-by-step human approval to broad goal-based execution, allowing agents to run continuously on recurring workflows such as following up on stalled sales deals, summarizing regulatory changes, and processing purchase orders.

The update also adds multi-dataset analytics, letting users query across disparate data sources — including Snowflake and relational databases — in natural language without needing to pre-join or technically prepare data. Quick inherits metadata context from existing data catalogs like AWS Glue, Databricks Unity Catalog, and Collibra, and enforces existing access permissions through identity propagation. A redesigned activity feed lets users approve requests, reply to emails and Slack messages, and prioritize updates via thumbs up/down — all within a single conversational interface.