Skip to the content
AI and Cybersecurity Learning and Practice Center Test automation Open the Learning Center

WeaknessASC-2026-0066Other weakness

Prompts and outputs logged with sensitive data

MediumMitigated

Observability tools, proxies and debug logs store full prompts and responses, including personal data and secrets, often with broad access and long retention.

How to detect it

  • Full prompt logging enabled in tracing or gateways
  • Logs readable by many teams or third parties

How to resolve it

  • Redact personal data and secrets before logging
  • Restrict access and set retention on AI logs

How to investigate

  1. Find where prompts are stored and who can read them
  2. Purge sensitive records

Sources

Open it in the Learning Center's vulnerability database