Article
Deep Dive into OWASP LLM Top 10 and Prompt Injection
Paul Duvall · paulmduvall.com
A practitioner's walkthrough of how injections actually happen in RAG and agent apps.
The question it answers
Read with it
- The primary source: API data is not used for training by default, with retention controls explained. openai.com Guide Enterprise privacy at OpenAI OpenAI
- The developer-facing detail on what is stored, for how long, and how to reduce it. developers.openai.com Guide Data controls in the OpenAI platform OpenAI
- The policy to quote when a customer asks whether their data trains the model. openai.com Guide How your data is used to improve model performance OpenAI
- Anthropic's equivalent answer, including the commercial-products carve-out. privacy.claude.com Guide Is my data used for model training? (Anthropic Privacy Center) Anthropic
- Anthropic's own account of what its API keeps: your prompts and Claude's outputs are not retained by default, with zero data... platform.claude.com Guide API and data retention (Claude docs) Anthropic
- Where to pull compliance artifacts when an enterprise customer sends a security questionnaire. anthropic.com Guide Anthropic's Transparency Hub Anthropic
eChai Partner Brands
The partnership →
