Skip to content
Sırdaş
es · Español
Try it freeToolsAI datasets NewDevelopersServidor MCPCLIPythonn8nTypeScriptAgent SkillSkill: invoicesSkill: contractsGatewayRecipesKYB package reviewBlogBusiness

Shadow AI document risk check

Ten questions about how documents with client, patient or financial data reach AI tools in your organization today: chat accounts, your own apps, coding agents and online converters.

You get an exposure level and the fixes your answers point to, in order. Some of them are Sırdaş tools; most are policies and contracts that no tool replaces.

A self-assessment to start the conversation with your team, not an audit or a compliance opinion. Reviewed:

  1. Do these documents include regulated data (health records, SSNs, bank or tax data)?

  2. Do people on the team paste client or patient documents into consumer AI chat accounts?

  3. Is there a written rule for which data may go to AI tools, and do people know it?

  4. Does your AI vendor contract exclude training on your data and set a retention period?

  5. Is personal data removed before documents reach a model?

  6. Do your own applications send customer documents to an LLM API?

  7. Is there a log of which kinds of data were sent to AI, by which app?

  8. Do people upload client files to free online PDF converters?

  9. Do coding or desktop agents (Cursor, Claude Code, Copilot) have access to folders with client documents?

  10. Could you list what was sent to AI services last month if a client asked?

0 of 10 answered

Your answers stay in this page's memory: nothing is sent or saved, and reloading clears them.