Case 85BFC3 · AI security · L3 Applied

Fine-Tuning vs RAG for Sensitive Data

Practise as: InterviewLevel: L3 Applied

Interview questionA team wants the model to answer from confidential internal documents. From a security view, should they fine-tune or use RAG?

  1. 01 Answer
  2. 02 Example
  3. 03 Trade-offs
What a strong answer covers

Try it out loud first. Then check yourself:

  1. Fine-tuning bakes data into weights: no per-user access control, and the model may regurgitate it to anyone who asks.
  2. Deletion is hard: removing one record means retraining, which clashes with erasure duties under DPDP or GDPR.
  3. RAG keeps data in a store you control: enforce user ACLs at retrieval, update or delete documents, cite sources.
  4. RAG brings its own risks: indirect injection through documents, poisoned or overshared sources, vector store access.
  5. Typical call: RAG for permissioned or changing data; fine-tune for style or format, on data everyone may see.

If the interviewer pushes back

  • They insist on fine-tuning for accuracy. What would you require before the tuned model ships?
  • How would you test whether a fine-tuned model memorised specific confidential records?

More AI security cards

All AI security cards →

cardosec draws a security topic and gives you a clock: explain it out loud with no notes, then see what you covered and what you missed. Free during early access.

Reviewed 2026-10-09