AWS has added live queries to AI-built apps in Amazon Quick, replacing frozen data snapshots with current results from governed QuickSight datasets. Each viewer’s existing row- and column-level permissions apply to the query, so the app can show different people different slices of the same data.
AWS AI Watch analysis
What happened
AWS says builders can describe an app in natural language; Quick’s agent discovers relevant datasets and writes the SQL while building it. After publication, the app runs that SQL when a viewer opens it, rather than displaying numbers captured when the app was made. The feature supports SPICE and Direct Query datasets.
Each viewer must be an authenticated Quick user and give consent for a dataset the first time it is used. Builders must approve the datasets found during app creation. AWS says Direct Query datasets from different data sources cannot be combined in one app, and a SPICE dataset shows refreshed data only after that dataset is refreshed. Live queries also have query and result-size guardrails: when results exceed what the app can carry, it asks users to narrow the query rather than showing an incomplete result.
Why it matters
An app built to answer a question about sales, support tickets or renewals can now query the underlying governed data when someone opens it, instead of relying on a snapshot that may already be out of date. The permissions matter just as much as the freshness: AWS says queries run as the viewer, with existing row- and column-level security applied.
That could spare teams some manual refreshes and custom data plumbing. It is not a free pass around setup: viewers need Quick accounts, dataset consent is required, and Direct Query has a same-source limitation. AWS describes the capability, not measured time or cost savings.
Our read
The useful idea is giving an AI-built app live access without asking organisations to invent a new permission model for it. Teams with governed QuickSight data can now try building an app in plain language and sharing it with authenticated Quick users. Before rolling one out, check which datasets it uses, who can see their data, and whether SPICE refresh timing meets the need. Fresh data is only useful if it is fresh in the right way for the people seeing it.
What to watch
- Whether live queries and per-viewer permissions behave as intended in real shared apps.
- How often SPICE datasets need refreshing for teams that need current figures.
- Whether AWS expands support for combining Direct Query datasets from different sources.
Discussion spark: Would you let an AI-built app query governed business data live for each viewer, or would the consent and setup requirements make you stick with static reports?
Sources and evidence
- Serve live, governed data in AI-built apps with Amazon Quick (1 October 2026, 19:49 UTC)
not affiliated with or endorsed by Amazon Web Services (AWS)