Small AI database questions can become big scans

A small AI-generated database query, such as "Show me customers at risk," can unintentionally trigger a large, resource-intensive scan if the agent joins multiple tables and retries failed queries. To prevent performance issues, row limits on production database servers should be enforced as a safety boundary rather than a user interface preference. Additionally, AI models should never summarize a limited preview of rows as if they represent the entire database.

A small question can become a big database scan when an AI agent writes the query. “Show me customers at risk” sounds harmless. But depending on schema context, the agent might join: Then it may retry when the first query does not answer the question. For production MCP database servers, row limits should be treated as a safety boundary, not a UI preference. Useful defaults: Longer version: Row limits for AI database agents The model should never summarize 50 preview rows as if it saw the whole database.